First published in 2008 by No Exit Press, P.O.Box 394, Harpenden, Herts, AL5 1XJ www.highstakespublishing.co.uk
© Rick...
631 downloads
5710 Views
418KB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
First published in 2008 by No Exit Press, P.O.Box 394, Harpenden, Herts, AL5 1XJ www.highstakespublishing.co.uk
© Ricky Taylor 2008
The right of Ricky Taylor to be identified as the author of this work has been asserted in accordance with the Copyright, Designs and Patents Act 1988. All rights reserved. No part of this book may be reproduced, stored in or introduced into a retrieval system, or transmitted, in any form or by any means (electronic, mechanical, photocopying, recording or otherwise) without the written permission of the publishers. Any person who does any unauthorised act in relation to this publication may be liable to criminal prosecution and civil claims for damages. A CIP catalogue record for this book is available from the British Library. ISBN 978-1-84344-054-3 2 4 6 8 10 9 7 5 3 1
Printed and bound in Great Britain by Cpod, Trowbridge, Wiltshire
This book is dedicated to Jo, Josh and Isaac
CONTENTS Introduction
7
Chapter 1 Why use a betting system?
14
Chapter 2 Developing systems
21
Chapter 3 Form figures
30
Chapter 4 The claiming jockey system
43
Chapter 5 A classy system
53
Chapter 6 Betting market systems
62
Chapter 7 Pedigree Profits
87
Chapter 8 Trainers to follow
109
Chapter 9 Systems for all-weathers
128
Chapter 10 Making a profit from negative factors
136
Chapter 11 Betting systems
142
Selected bibliography
157
Introduction Can you beat the bookies with a system? There are plenty of people who will say that you can’t, but I can tell you that the answer is definitely yes, and I have developed scores of profitable systems to prove the point. However, I’m not the only punter who makes their racing pay by following systems. I know of numerous professionals who adopt a systematic approach. The most successful and the most sophisticated of them all is probably a guy called William Benter. He runs a fully automated, computerised betting system in Hong Kong that makes a return of millions of dollars per year. He is living proof that you can beat the races by following betting systems, and thus make a decent living from doing so.
What, though, is a betting system? There isn’t an entry in the Oxford English Dictionary to define the term but, as someone who has spent the best part of twenty years researching and betting on systems, I would say that a good betting system is characterised by three things. It has to be logical, profitable, and above all it needs to be inhuman. In my view it is the latter point that is the most important, because to me a system isn’t such unless it removes the human element as far as possible.
In Ricky Taylor’s betting dictionary a system is defined as the mechanistic application of a set of rules to predict an event, which is then bet on. The research behind the system may be the result of painstaking human endeavour, but its application will be robotic. Each day’s race cards will be systematically assessed, with the system applied in the same way to each horse in every race. There is no place for human judgement when applying
PROFITABLE BETTING SYSTEMS the system. Once human judgement is involved it ceases to be a system. This might sound a bit dogmatic but humans are hopeless at making the kinds of decision important to picking winners, i.e. deciding on the most probable outcome. Probability isn’t something that humans are programmed to calculate.
We humans are terrible at weighting the importance of different pieces of information, and too prone to thinking that something is likely to happen simply because we think it will happen. We often believe that a horse can win a race because we have focused on just one positive aspect of its form, and have ignored all the negatives. We might have bet our hard-earned cash because we liked its jockey or because our favourite tipster had tipped the wretched beast, or it had the name of someone we knew. We make bets for lots of different reasons but seldom on the basis of the cold facts.
Betting systems are different. Good systems only make selections according to the data and are not influenced, like we are, by irrelevant factors. They don’t have good or bad days. They are consistent and this is why I believe that it is only through the research, development and implementation of betting systems that a punter can win in the long run.
Betting systems have fascinated me for about as long as I can remember. My Dad is probably to blame. His lifelong interest in horseracing and betting systems was bound to rub off on me, and I recall that one of the first books that I ever read was one that I borrowed from him called The Red Wizard betting formula. This was a system that made winning sound so very easy. You simply had to look up the runners and riders from a newspaper, apply the ‘formula’, and hey presto you had found the winner. I was hooked! After that I think I developed my first system when I was about nine-years-old. I
8
INTRODUCTION remember it being fantastically successful with its first half-dozen selections, and I started to dream of early retirement before I left primary school, but needless to say the early winners proved to be a flash in the pan. However, I wasn’t put off and persevered with developing more systems, from dust-mite invested back copies of the Racing Post. Schoolwork was put on the back burner. Form study was much more interesting.
Then, when my parents purchased me my first computer at about age 11, my immediate thought was ‘will this help me to pick more winners?’ I spent hours bent over that wretched machine trying to find out the answer to that question. This was later to prove to be time well spent. The computer allowed me to develop more sophisticated systems, and by my teenage years I was a geek who was starting to supplement my pocket money with profits from systems I had developed. I recall that one system that I was running at that time managed to select a 33 to 1 winner in a six runner race, and another impressed schoolmates and teachers alike by picking Quest for Fame to win the Derby at 9 to 1.
My obsession with betting systems has never left me, and it certainly is an obsession. I have no doubt spent more time than is healthy trying to work out the answer to the perennial question of ‘which horse will win the race?’. Indeed I have often wondered what I would have achieved if I had devoted more of my time to more socially useful activities. My family and friends would probably say not a lot, and they are no doubt right. I have, however, derived a lot of pleasure from developing and running betting systems.
It is probably the background research into a system that fascinates me the most. I love finding out new winning angles, partly because of an in-built drive to seek out new facts and partly through sheer greed. In a way I suppose my
9
PROFITABLE BETTING SYSTEMS search for successful betting systems is similar to that of the alchemists and their quest to turn lead into gold. However, I can probably claim more success. I have actually been able to develop a range of systems down the years that have proved to be consistently profitable, and many of them are detailed in this book.
I was motivated to write this book for a variety of reasons. Primarily I started writing because of my love for horseracing and betting systems and because I wanted to share some of my research with others. I was also fed up with reading the same old tired books on horse race betting systems. You know the type, the ones that assert that you should always back a horse wearing blinkers for the first time, or one that has just changed trainers or is making its first appearance after being gelded. These books do not cite any statistical evidence in support of these ideas, or if they do it is merely a handful of carefully selected examples. I know plenty of fellow punters that buy these types of book and slavishly follow the recommended systems and half-baked staking plans, and they are all broke.
I wanted to write a book that was different. I wanted to write a book that examined betting systems from a scientific point of view, by testing different types of selection method against the available data. In part, I was inspired by Nick Mordins ‘Betting for a living’ which was probably the first book on British racing that tried to look at betting systems in this way. I was also influenced by a book I picked up from the United States by William Querin called Computer discoveries in thoroughbred racing. In this book the author analysed the results of thousands of races run in the US in the late 1970s and came up with some fascinating winner finding systems. It is rather dated now (although some of its findings have stood the test of time) but it is well worth a read.
10
INTRODUCTION In this book I have tried to follow in the footsteps of Mordin and Querin by approaching the study of betting systems in a scientific manner. The systems contained in this book are therefore based on painstaking research on the results of thousands of races run in Great Britain and Ireland over the last dozen years or more, covering both flat and National Hunt racing.
I have structured the book mostly around my favourite betting systems. However, in Chapter 1 I have tried to further develop the arguments I have presented here as to why betting systems are the best way of making your racing pay, drawing on research from cognitive science and other disciplines. Given that you have purchased this book I am probably preaching to the converted but I make the case because I know that so many of our fellow punters are cynical about the approach. In Chapter 2 I move on to discuss how you can go about developing your own systems, and identifying duff ones.
In the rest of the book I then discuss some profitable systems and strategies, starting with form figure based systems in Chapter 3. I recall that one of the first systems that I ever read was based around such a method. That system wasn’t the dud I thought it was at the time and I have made a steady income stream from betting on horses simply because they have recorded certain form figure combinations.
I also like to follow certain stables and a number of the systems described in this book are based on following particular trainers in certain types of races. In Chapter 4, I detail a nice little system that is based around backing three and five pound claiming jockeys when they are been given the leg-up by certain trainers.
11
PROFITABLE BETTING SYSTEMS A great deal of ink has been expended on the topic of a horse’s class, with every book on form study claiming it to be an essential factor in picking winners. In Chapter 5 I don’t disagree and describe a fairly simple system based on official handicap ratings that turns in a nice little profit year after year.
There are a number of books on horserace betting that describe what are termed ‘technical’ systems. These types of system are not based on form factors. They ignore the trainer, the jockey and the horse’s previous performances. They are based on the betting market. These types of systems are usually hideously complicated and require an honours degree in mathematics to comprehend. I am told that they do work but I haven’t got a degree in maths and so I ignore them in the main. However, in Chapter 6 I do discuss a number of simple systems based on the betting market, particularly the circumstances when it is profitable to back and lay the favourite, which I hope will be useful to those of you that like to back and lay on the exchanges.
In Chapter 7 I get back to fundamentals by considering how pedigree analysis can be put to good use. Most punters ignore a horse’s bloodline when studying the form either because they do not believe the information to be important or because they cannot obtain basic information on a horse's sire, dam and grandsire. This creates an opportunity and over the years I have made good money by studying a horse’s breeding, and by developing systems around this theme.
As I said earlier I spend a great deal of time researching the performance of certain trainers and identifying circumstances when it is profitable to follow them. In Chapter 8 I therefore give a number of systems based on trainers to follow, for flat and jump racing.
12
INTRODUCTION
The basics of the racing game never change but the sport itself is constantly evolving. One of the biggest and most important changes in recent years has been the advent of all-weather racing. This form of racing has been kind to me and in Chapter 9 I share with you a number of systems for polytrack and Fibresand fare.
When it comes to studying racing form most punters try to eliminate those horses with big negative factors against their name. For instance, a horse wearing blinkers or a tongue strap for the first time can often be quickly eliminated because these factors are very big negatives. This makes a lot of sense because statistically horses with these characteristics seldom win. However, they do sometimes win, and often at fantastic prices. In Chapter 10 I describe a few ways in which you can identify horses that at first sight seem to have little chance but actually represent good betting opportunities.
Over the years I have been asked various questions in relation to developing and implementing betting systems. In the final Chapter I attempt to answer some of these, and take the opportunity to pass on a few other facts that you might find interesting.
I hope that you enjoy reading the results of my research, but above all I hope that the systems described prove to be as profitable for you as they have been for me.
13
Chapter 1: Why use a betting system? I have an academic background in psychology and the one lesson that my undergraduate course taught me was that human beings are absolutely useless at making decisions. We Homo sapiens simply do not have the faintest idea how to process information in an objective and systematic way. This makes us useless when it comes to working out the true odds of something happening. A good example of this comes from surveys of lottery players. They all believe that the odds of winning the jackpot are less than a thousand to one against, and are amazed when told that the true statistical odds are about 14 million to 1!
Human beings are particularly bad when we are confronted with a range of information upon which to make a decision. In these circumstances we tend to place too much emphasis on one piece of evidence and not enough on others. In some instances we actually ignore key pieces of information because we do not understand them, or because they don’t fit with our preconceived theories about how the world works.
We are also heavily influenced by the source of the information, believing that a piece of information must be true because we like the look of the person saying it or the sound of their voice. Indeed research shows that people are more likely to believe someone speaking in a Yorkshire accent than anyone else, regardless of whether they are speaking the truth or not. Apparently Yorkshire folk are viewed to be more trustworthy (which will please my fatherin-law) but on the grounds that they are less intelligent and therefore less cunning (I haven’t mentioned this to him!).
WHY USE A BETTING SYSTEM Two psychologists called William Grove and Paul Meehl produced the most comprehensive research on the decision-making abilities of human beings that I have found. They set about assessing the decision-making ability of professional experts, working in fields as diverse as criminal justice, clinical psychology and education by comparing the accuracy of their decisions to those made by statistical models.
The term ‘statistical model’ implies a high level of sophistication, and one assumes that models require a high level of computation and an honours degree in mathematics to comprehend. However, the models reviewed by Grove and Meehl were remarkably simple systems. In many instances they required only two or three system inputs. In the case of one model, that tried to predict whether or not an offender would reoffend on release from prison, the user only had to input the age of the offender and the offender’s number of previous convictions. Points were awarded for offenders of a certain age and additional points were awarded for the number of previous convictions. A simple rule was then applied that stated that if the offender had more than a given number of points then they were more likely to reoffend than an offender with fewer points.
Grove and Meehl reported on a study based on 3,000 criminal offenders given parole, which compared the predictions of the statistical model to the expert opinion of three highly experienced prison psychiatrists. The latter were salaried to make decisions about which offenders were safe bets not to reoffend and which were not. The results were unambiguous. The model, despite is simplicity, proved to be much more accurate in its predictions than the psychiatrists! In fact you would have been better off tossing a coin to decide which offenders should have received parole rather than rely on the expert opinion of the shrinks. It would certainly have been more cost effective!
15
PROFITABLE BETTING SYSTEMS In a further study Grove and Meehl reported the results on an experiment that compared a system to the expert opinion of college tutors in predicting academic grades for a large group of undergraduates. In this instance the college tutors thought that they had a great advantage over the model in that they not only had access to the two pieces of information used by the model (both known from previous research to be predictors of college academic grades), but also had access to a good deal of additional information that one would usually consider relevant. This supplementary information included data on students IQ, previous academic record, and a written report from each student on their academic and vocational interests. In addition the college tutors also had the opportunity to interview the students prior to making their grade prediction. However, despite being in receipt of all this additional information the college tutors backed more losers than winners. Their grade predictions were greatly inferior to those predicted by the statistical model!
This finding is interesting because it shows how difficult we humans find it to process large amounts of information and to weight the importance of each item of information accurately. The college tutors simply had too much information at their disposal and got distracted by irrelevant data, or failed to appreciate the full importance of key information.
Grove and Meehl, after reviewing the results of these two studies and a further 134 studies, across a range of professional fields, concluded that it was clear that even crude statistical models (or what I call systems) are superior to expert opinion in making probabilistic judgements. In their view, when making an odds call about whether or not an offender will re-offend, or the likelihood of a student attaining a certain grade, you would be better off relying on some well researched statistical model or system than the judgement of human experts.
16
WHY USE A BETTING SYSTEM This is a lesson not lost on the insurance or financial sectors. The insurance industry has for years been using what they call ‘actuarial models’ to judge whether someone is a good insurance risk or otherwise. These actuarial models are basically betting systems for working out whether or not you are a good insurance risk. You may not know it but these systems are used all the time. For instance, every time I apply for my car insurance to be renewed I am asked questions about my past form. I’m asked my age, my driving experience, my address, whether I store my car in a garage or not, whether or not I have had an accident or had a car stolen etc. The answers I give to these questions all feed into some sort of statistical model that computes the probability of me making a claim on my insurance policy, and the level of probability determines the level of the premium that I will pay. Should the model conclude that I am a safe bet then my premiums will be low, but if I am deemed odds-on to make a claim then my premiums will be high or the company may refuse to insure me because I’m judged to be too bad a driver.
Similarly credit card companies ask for a whole range of information upon which to base a credit assessment. The type of information that they use to make their prediction is not determined at random. A huge amount of investment is made into researching which items of data best predict whether or not I’m likely to default on a loan.
You may well wonder what on earth this has got to do with using betting systems to pick winners. My answer is that if human experts can make the wrong odds call in a range of fields as diverse as education and psychiatry then they can also be badly wrong when it comes to betting on the horses. This is fortunate for bookmakers – the more rational member of the human species – and explains why for every pound wagered on the favourite the
17
PROFITABLE BETTING SYSTEMS bookie returns less than 90p to the punter. This salutary statistic demonstrates the challenge facing punters in their bid to make the racing game pay.
In my view most punters lose because they do not appreciate a horse’s true odds of winning. In other words they are prepared to back an even-money favourite when in fact the actual probability is considerably less than a 50-50 bet. This comes back to the point made by Grove and Meehl that we humans are not good at working out probabilities because we do not systematically review all the evidence available, and the information that we do process we weight inappropriately. This definitely applies to betting on the horses. You only need to open up the form pages of the Racing Post or the racing pages of any national newspaper to be confronted with masses of information and opinion on which horse will win a race. The sheer volume of information and opinion makes it difficult to work out what evidence is worth considering and what isn’t.
All of this data and opinion is not of equal value. In a statistical sense some pieces of information are closely related to a horse’s true odds of winning while others are totally unrelated or of minimal relevance. However this begs the obvious question of, which is which? In my firm opinion the answer to this question can only come from a systematic statistical analysis of all the key variables in the form book, followed by use of the results of this analysis to generate objective betting systems.
This cold, calculated approach can be the only path to success because it takes out the human element as far as possible! This is why insurance and credit card companies invest heavily in developing the equivalent of betting systems to work out our chances of crashing our car or defaulting on a loan. In my view this is a fact that shouldn’t be lost on punters. If multi-billion pound
18
WHY USE A BETTING SYSTEM industries trust systems more than subjective human opinion in making decisions then punters should take notice. The successful punter will take heed and will spend his or her time developing and using systems rather than trying to work out the winners for themselves.
If you are a systems purist like me the ideal scenario would be one whereby you collected all the necessary data needed to apply a particular system, worked out which horse qualified under the system and then placed a bet, the size of which would have been carefully calculated. In this approach one isn’t distracted by the fact that your mate down the pub knows a guy, who knows another guy, who thinks such and such a horse is a good thing in the 4.30 at Kempton Park.
In the systematic approach you know that your bet has been arrived at by a careful statistical analysis of the all the key data, and that the odds are in your favour. The level of your stake has been carefully calculated. It has not been determined by personal psychological factors such as you are having your tenth consecutive losing bet and your confidence is shot to bits or (and this is much worse) you are on a winning streak and feel that you can do no wrong. The systematic approach safeguards you against all these emotions. It protects you from making that one off bet that was out of proportion to every other bet you ever made, or from only staking two quid on that 33 to 1 winner you had the other day when your normal stake is a score. This is why betting systems are the only way to make your racing pay. They protect us from ourselves.
Not everyone believes in betting systems like I do. I know that Nick Mordin in his book Winning Without Thinking said that he doesn’t use betting systems is a rigid manner but instead prefers to use them as an aid when making a
19
PROFITABLE BETTING SYSTEMS selection. I should say that I disagree. I’m a systems purist and prefer only to stick to the rules of a system, regardless of any other considerations. I’m not into making subjective judgements. Boring? Not at all. Excitement comes from winning.
20
Chapter 2: Developing systems It is great fun to develop your own betting systems but it can be very time consuming if you don’t have the aid of a computer. In the past, before electronic formbooks became readily available, I used to develop my systems from newspapers. I had a large wardrobe that was stuffed from top to bottom with old, yellow, dust mite infested back issues of the Racing Post. When I had an idea for a system I would then test the system by wading through all my back issues and recording all the results. That was painstaking work. I’m glad those days are gone, and my wife is probably even more relieved!
With the advent of computer databases it is possible to develop and test systems in a matter of minutes and if you are really serious about betting systems then you really do need a computer and need to buy one of the many excellent databases that are available. I’m not going to recommend any particular package, because the field is evolving all the time, but I’ve used a variety of different packages over time. They all have their strengths and weaknesses and it is probably a personal decision about which you prefer. I prefer packages that allow me to download the raw data onto statistical software like Microsoft EXCEL for analysis. Raceform Interactive or Dataform are two packages that I have tried that allow you to do this.
It is also ideal if you have some knowledge of statistics and some computer programming skills. I’m fortunate in that I have studied statistical and research methods and I’ve trained myself to use some fairly sophisticated statistical software programs. However, you have probably worked out by now that I’m a bit of a systems anorak, and most normal people wouldn’t want to go to these
PROFITABLE BETTING SYSTEMS lengths to develop betting systems. In this Chapter I therefore want to give you a few short cuts to developing robust and reliable systems. You will then avoid some of my early and costly mistakes.
Data
The first thing you need to do when developing betting systems is to get hold of as much data as possible. In the modern world this is fairly easy. There are tonnes of data on the Internet and you, as discussed above, can purchase electronic formbooks that allow you to download past racing results onto spreadsheet packages for further analysis. There are also menu driven software packages that allow you to develop and test systems without having to bother to learn how to program a computer. Some of this stuff isn’t cheap but data and software are the fundamental tools of the trade. I look at systems development as a business and all businesses have set-up costs. The purchase of a computer, data and software are your set up costs. Once you have developed your first profitable system you can soon recoup your investment!
I wouldn’t try to skimp on paying for data. You need samples that run into tens of thousands in order to generate genuinely valid and reliable systems. A computer with plenty of memory and processing power is therefore essential. I once made the mistake of buying a cheap computer that was under-powered for processing the size of database that I had put together. It took the machine an age to process all the data and it wasn’t long before the useless thing burnt out. However, nowadays even the most basic PC is capable of processing a decent sized database in a few seconds.
22
DEVELOPING SYSTEMS Ideas for systems
Once you have your data in place you can then start to develop some ideas for a system. Nick Mordin has written a lot about this and in the bibliography at the end of this book I have highlighted a couple of his books that cover systems development. The most relevant is probably his ‘Winning Without Thinking’. The key point that he makes in this text, and one that I very much agree with, is that you have to be original when developing systems. If you simply reinvent systems that have been used before then you are unlikely to make a profit because, if it is a genuinely good system, then the factors on which it is based will already be incorporated into the betting odds. What you should be striving for is an angle that no one has really researched before, and is probably not reflected in the betting odds. In this instance you are much more likely to be able to make a long-term profit. This doesn’t necessarily mean that you need to develop very complex systems. I would always advise that you should keep it simple and focus on horse race fundamentals, namely the horse’s ability, jockey, trainer, pedigree, fitness and consistency, and the betting market. As you read on you will see that I have structured this book around these same themes because they provide a rich source of profitable systems.
Testing your ideas
Once you have your data, and you have some ideas for a system, then you can start testing them against the available data. This all sounds very logical but in reality there is a certain amount of circularity in developing systems. For instance, you may have an idea but cannot find the data on which to test it, and so you can’t go on to progress the system. Your system ideas therefore
23
PROFITABLE BETTING SYSTEMS need to be framed around the available data. The richer your data, and the more variables it contains, the more system ideas you will be able to test.
Once you have an idea, and the data to test it, you can then work out whether it is profitable or not. In the test phase you may find that your system isn’t profitable and so you may want to do more research on the systems variables to find the best combination, or add other variables to improve the results, and then to test it again.
However, one of my early mistakes was to develop a system in this way. I then rushed to implement it straightaway. This proved to be costly and I couldn’t understand why the system didn’t work in real time. After a lot of reflection I realised that what I had actually done was to back fit a system to past results.
In a back fitted system the selection rules are manipulated to account for a sample of previous results. For instance, if the system developer finds that his or her system picks a loser, the rules are then changed slightly to eliminate this selection. Similarly some rules are changed to accommodate a longpriced winner. This process is repeated until the system produces a respectable number of winners and a decent level of profitability on past results. This is what I did when I developed my early systems. I was using the same sample of data to build my system and to refine my ideas, and using that same data to test its validity. What I should have done was use what is called a split-half sample.
A split-half sample is exactly what it says on the tin. It is a sample of data that has been split into two halves. In one half of the data you develop your system. You can then play around with the variables to your heart’s content
24
DEVELOPING SYSTEMS until you have something that looks to be sensible and profitable. At this stage the system is a pure back fitted system. In order to test it properly you then run the system against the ‘unseen’ data in the other half of the sample. This is what is called the validation sample. If the system shows a profit over both samples, and provided that the samples are large enough, then you can be fairly confident that you have found a genuinely profitable system.
When looking at commercially available systems the sign of a back fitted system is a long list of complicated selection criteria, which often appear illogical. They tend not to work when applied in real time because they are not based on sensible, proven form factors. A good system, like the ones described in this book, are constructed around fundamental form factors that most professional punters would regard as important.
Racing logic
The above brings me to my next point about betting systems. Regardless of any other consideration, they must conform to racing logic. It is vital that the variables used in a system are sensible.
I once stood in a betting shop in open-mouthed amazement when a fellow punter punched the air to celebrate a winner and turned to me and said “First letter ‘R’ system. Never fails”. I think most right thinking people would agree that a horse with a name beginning with the letter ‘R’ is no more likely to win than any other horse with a name beginning with any other letter of the alphabet. But my betting shop colleague wasn’t convinced by this argument. He simply couldn’t see that there was no logic to his system. After all, as he made clear, he’d just had a 20 to 1 winner and I hadn’t!
25
PROFITABLE BETTING SYSTEMS Nevertheless I prefer to base systems on the basics of the game. It is better to stick to variables that most racing professionals would agree are important to winner-finding. You can then combine these in unique ways, or look to measure a key form concept like no one else. A good example of the latter comes from Andy Beyer’s work on speed figures in the United States. This brilliant form guru made a fortune in the 1970s and early 1980s by assessing horses’ ability using speed figures. No one was much using speed figures at the time and, compared to those who did, Andy had found a unique way of calculating them. He later sold the method to the Daily Racing Form (the US equivalent of the Racing Post) but not before making a fortune from his figures. Objective variables
When developing systems you should only use quantitative variables. In other words only use factors that can be measured in a consistent and objective manner. Avoid at all costs systems that rely on qualitative information. I know of many systems that use qualitative data on horses’ looks. This is based on sound racing logic that good-looking horses are usually the most able because they have a good physique, and a high level of fitness. The problem is that what one-person judges to be a good-looking horse another person will judge to be a pork chop. As an example I remember reading in Pat Taaffe’s excellent autobiography My Life and Arkles , a story about Lord Bicester who always purchased what he considered to be very good-looking horses. When his horse Royal Approach won the Irish Grand National a former leading jockey patted the Lord on the back to congratulate him and said “Good horse, but not much of a looker is he?” to which the Lord replied “when was the last time you looked in the mirror”! It is therefore better to steer clear of opinion
26
DEVELOPING SYSTEMS and focus only on those variables that you can measure in a consistent and reliable way.
Look for consistent profits
A number of so-called profitable systems do not record consistent levels of profit, year in, year out. They often record a lucky year in which they make an extraordinary profit, which disguises the fact that in a normal year the system makes a loss. When testing systems it is therefore important to break down the results year by year (or month by month) and note the level of profit recorded in each period. I am always much more confident of a system that has shown a consistent profit over time. I’m particularly interested in whether the system shows a profit across years in its validation sample. If you find a system that shows a consistent profit in its validation sample then get your betting boots on!
Implementing the system
The whole point of using betting systems is, as I tried to explain in Chapter 1, to eliminate as far as possible the human element. However, this is easier said than done and it is very easy to break the rules of a system especially during a long losing run. In this instance the thing to avoid is reducing stakes because, as sure as eggs are eggs, when the run ends you will have your smallest stake on the long priced winner that would have got you into clear profit.
Another thing to avoid, when running a system, is changing the system rules as you go along. For instance, when faced with a losing run it might be tempting to tighten up on the systems criteria in the hope that this will increase
27
PROFITABLE BETTING SYSTEMS the chance of a winner. This, without detailed further analysis, could fundamentally alter the working of the system and destroy any chance of making a long-term profit. Provided that your original analysis was sound when you developed and tested it, then you need to stick with the rules of the system. If the system works it will come good in the end.
However, a big issue for system followers is knowing how long to stick with a system that is having a losing run? All systems, simply due to random statistical variation, will have losing runs. The question though is when is a losing run the result of bad luck or something more fundamental? I should confess that when I first started playing systems I was probably too quick to abandon genuinely profitable systems just because they had a bit of a losing run. Nowadays my rule of thumb is that if a system has a losing run 20 per cent longer than it recorded during its test period (and that the test included at least 100 selections) then there is probably something wrong with the system.
Another issue that you will encounter when you run systems is that your once profitable system will, for no obvious reason, turn into a loss maker. This could be the result of poor system development but, provided that you have followed the principles outlined above, could be the result of what I call system rot.
System rot is a problem to which I made reference to earlier, but not by name. It basically means that if a system is very successful, and it becomes well known, then it will end up being incorporated into the betting odds. The price of the system’s selections will consequently contract and the value that the system exploited in order to return a profit will have gone. A once profitable system may now become a loss maker. The only way to avoid system rot is to be discreet about your successful system. Keep it to yourself. This is easier said than done but try to follow the slogan of a World War II poster I once saw.
28
DEVELOPING SYSTEMS It read ‘Like Dad keep Mom’. You can, however, still give the game away by staking too heavily on the system’s selections. In weak betting markets you can contract the odds simply by the size of your bet. This isn’t a problem though if you take a price, but requires that you are able to get on.
In conclusion I would say that I have developed my most profitable systems by adhering to the following principles: x
The rules of a system should be based on sound racing logic.
x
The variables used should show some degree of originality.
x
It should be based on objective variables that can be measured. It shouldn’t use subjective data like opinions on the look of a horse.
x
The system must be developed and refined on a large sample of past data in order to produce robust results.
x
It should be tested on an equally large sample of validation data. In other words its performance should be tested on data that was not used to develop the system.
x
The system should show a profit in every, or nearly every, year on both the development and validation samples.
Provided that you follow these principles then I am confident that you will develop plenty of good systems that will reward all your hard work. My hope is that the rest of this book will spark your imagination and provide you with plenty of ideas, because it is imaginative and original systems that make the most money.
29
Chapter 3: Form figures There are a great number of betting systems on horse racing that are based around a horse’s previous finishing positions. I recall that one of the first systems that I ever read was based around such a method.
The system cost me twenty-five quid and it involved awarding points according to where the horse had finished in its last two races. As you would expect, the maximum number of points were awarded to the horse that had finished first in its last two starts. A horse could only score points if it had finished in the first four. No other form factors were considered.
The simplicity of the system caused me concern. For instance, it didn’t take into account whether the horse had finished first in a two horse race or had been placed fifth in a thirty runner handicap. Furthermore, it didn’t consider the class of the horse’s previous races, and so you could in theory award a maximum number of points to a horse that had won two sellers at Southwell on its most recent starts and was now running at Royal Ascot.
The system also didn’t consider the distance of the race, and whether the horse was now racing over a distance radically different from what it had done before. It also didn’t take account of the horse’s jockey or trainer, and it paid no consideration to the going, or the number of days since the horse previously raced. I therefore concluded that the system was far too simplistic to be worth following, and decided to cut my losses. I was already a pony down and didn’t fancy losing any more.
FORM FIGURES However, more recently I revisited my original form figure system and tested it to see how it performed over a large sample of previous races. I was surprised to find that the method did produce reasonable results. You wouldn’t have made a profit, but you would have done a good deal better using this system than if you had made selections at random. This led me to look again at systems based on form figure combinations. I was now curious to see whether it was possible to make a profit from such a simple approach.
The merit of recent finishing positions
The first thing that I analysed was the statistical probabilities of a horse winning according to its finishing position on its last, second from last, and third from last run for all flat races run in Great Britain in recent years. The results of these analyses are presented in Tables 3.1, 3.2 and 3.3.
Table 3.1: Finishing position on last start and % win on next start Finishing pos. Winners Losers Total % Win First
5,384
26,412
31,796
16.9
Second
5,428
26,072
31,500
17.2
Third
4,125
27,122
31,247
13.2
Fourth
3,265
27,191
30,456
10.7
Fifth
2,657
26,270
28,927
9.2
Sixth
2,081
24,579
26,660
7.8
Seventh
1,669
22,470
24,139
6.9
Eighth
1,409
19,977
21,386
6.6
Ninth
1,081
17,248
18,329
5.9
Tenth +
3,546
70,424
73,970
4.8
30,645
287,765
318,410
9.6
Total
31
PROFITABLE BETTING SYSTEMS Table 3.2: Finishing position on second from last start and % win on next start Finishing pos. Winners Losers Total % Win First
4,141
25,602
29,743
13.9
Second
4,229
25,243
29,472
14.3
Third
3,546
25,385
28,931
12.3
Fourth
2,937
25,057
27,994
10.5
Fifth
2,440
24,000
26,440
9.2
Sixth
2,043
22,131
24,174
8.5
Seventh
1,750
19,923
21,673
8.1
Eighth
1,424
17,616
19,040
7.5
Ninth
1,120
15,208
16,328
6.9
Tenth +
3,951
60,753
64,704
6.1
29,924
291,596
321,520
9.3
Total
Table 3.3: Finishing position on third from last start and % win on next start Finishing pos. Winners Losers Total % Win First
3,434
23,938
27,372
12.5
Second
3,478
23,798
27,276
12.8
Third
3,061
23,609
26,670
11.5
Fourth
2,628
22,955
25,583
10.3
Fifth
2,220
21,836
24,056
9.2
Sixth
1,893
19,957
21,850
8.7
Seventh
1,636
17,867
19,503
8.4
Eighth
1,403
15,680
17,083
8.2
Ninth
1,121
13,442
14,563
7.7
Tenth +
4,071
53,344
57,415
7.1
30,352
293,954
324,306
9.4
Total
The data presented in the above Tables reveal a number of factors that are worth discussing further.
I found it interesting that the finishing position recorded on a horse’s last start is more predictive of a win next time than the horse’s finishing position on its
32
FORM FIGURES second from last, and third from last start. For example, the win rate for those horses that won their last race was 16.9 per cent. But the win rate for those horses that won on their second last start was only 13.9, and the rate was a meagre 12.5 for those horses that recorded a win on their third from last start.
The system that I described earlier didn’t take heed of this fact because it weighted the finishing position recorded for a horse’s last and second from last start as being of equal importance. Clearly the predictive power of past form decays with time.
You probably noticed from Tables 3.1, 3.2 and 3.3 that second placing were more predictive of a win next time out than any other finishing position, including previous wins. This applies to whether the finishing position was recorded for the horses last, second or third from last runs. This makes nonsense of all those systems that award the greatest number of points for winning performances. The greatest number of points should actually be awarded for second places.
It is worth speculating why second places record a higher strike rate than previous wins. The most probable explanation is that last-time-out winners find themselves running against better class opposition the next time they race, either because the handicapper gives them a higher rating, or because the horse’s connections become more ambitious. Whatever the reason, previous winners are set a greater task the next time they race. In contrast the horse that finishes second tends not to experience such a significant change in its handicap rating. It can consequently compete next time against roughly similar opposition. In addition, the connections of runners-up have no good reason to raise their horse in class if it has already been beaten at a lower grade.
33
PROFITABLE BETTING SYSTEMS System development
In terms of system development one could try to use the probabilities presented in Tables 3.1, 3.2 and 3.3 to develop a revised scoring system that awarded the maximum number of points for horses that record three previous second places. This would be logical but it does not take into account the possibility that different form figure combinations may record a higher win rate. For instance, horses that record form figures of 112 may have a higher win rate than horses that have three seconds to their name. As a result I decided to research form figure combinations for different types of races rather than develop a revised point scoring system.
In my analysis I considered only a horse’s previous three finishing positions. I did this because the predictive value of form figures declines with time. For instance, as seen in Tables 3.1 and 3.3, a win last time out is more predictive of a win next time than a win that was recorded on a horse’s third from last run. The number of form figure combinations also becomes too large if one starts to consider more than three previous performances. I also reclassified form figures into the following six categories: x
First (denoted by the figure ‘1’)
x
Second (‘2’)
x
Third (‘3’)
x
Fourth (‘4’)
x
Fifth or any other placing (0). This category included horses that refused to start, were pulled-up or brought down etc.
x
No run (if the horse had not previously raced, or had raced only once or twice).
34
FORM FIGURES I categorised the data in this way because if one starts to analyse all placings the number of form figure combinations become too numerous.
The form figures were also derived on a first-past-the-post basis. In other words if a horse was disqualified and its placing changed by the stewards, I ignored the revised placing and used the original finishing order (i.e. as the horses passed the finishing post). I did this because I have often found that, in the event of intervention by the stewards, it is the original result that more accurately reflects the merit of each runner.
It should also be noted that I did not distinguish between form figures recorded across different racing seasons. Thus if I read in my racing paper the form figures ‘1-21’ I ignored the seasonal delimiter (-) and interpreted the combination as 121 (the figure on the far right being for the horses’ most recent start).
The results
In the tables that follow I present the results of my findings for different form figure combinations, for different types of races. Due to space I have only included those combinations that returned a profit, and I have only included those combinations that had a sample of at least fifty. This makes the results more reliable.
Non-juvenile stakes races
Non-juvenile stakes races record a total of twenty profitable form figure combinations. Table 3.4 shows that backing only those horses that recorded one of these combinations would have netted a return of just over 11 per cent.
35
PROFITABLE BETTING SYSTEMS Indeed some of the combinations record extraordinary profits. But don’t get excited. You have to be cautious about the bigger returns because they tend to come from the smaller samples. For instance, the combination 242 returns an 80 per cent profit but the sample is only 59. In contrast, the combination 120 returns just 2 per cent but on a sample of 371 races. This is probably the more reliable finding and one could be confident that this combination would return a profit in the future. Small returns though are not always associated with the larger samples. The form figure triple of 230 produces a rate of return of nearly 15 per cent and is based on a sample of 228 selections.
A placing on either the horse’s last, second from last or third from last run seems to be highly important in this type of race, with all of the 20 combinations recording at least one placing (either first, second or third). It is also interesting that 8 of the 20 profitable combinations show at least one first place. It doesn’t seem to matter whether the win was on the horse’s last, second from last or third from last run.
Table 3.4: Non-juvenile stakes races (non-maiden) Last
Second
Run
Last run Last run
Third
Wins Total %Win Profit/loss (£) Profit/loss (%)
2
4
2
10
59
16.9
47.50
4
1
4
13
71
18.3
23.00
32.39
2
2
4
22
77
28.6
27.38
35.56
3
3
2
19
80
23.8
3.71
4.64
80.51
1
4
2
19
83
22.9
18.05
21.74
2
2
3
28
85
32.9
15.90
18.70
3
3
3
16
86
18.6
3.78
4.40
3
2
3
16
88
18.2
13.85
15.74
3
2
4
25
90
27.8
45.78
50.87
2
4
1
17
93
18.3
9.09
9.77
2
3
2
20
94
21.3
9.82
10.44
36
FORM FIGURES Last
Second
Run
Last run Last run
Third
Wins Total %Win Profit/loss (£) Profit/loss (%)
2
1
4
23
102
22.5
3.30
3.23
4
2
1
23
121
19.0
27.39
22.64
1
1
3
41
165
24.8
0.76
0.46
1
0
4
45
223
20.2
9.62
4.31
2
3
0
38
228
16.7
33.74
14.80
4
1
0
43
239
18.0
14.15
5.92
3
2
0
43
279
15.4
2.40
0.86
0
3
2
50
324
15.4
14.84
4.58
1
2
0
89
373
23.9
8.78
2.35
600
2960
20.3
332.84
11.24
Total
Non-juvenile handicaps
This type of race records 13 winning form figure combinations. Again, as with non-juvenile stakes races, a recent placing seems to be important. The profits from this race category are a modest 4 per cent but the sample size is very large and as a result we can be pretty confident that these combinations will be repeated in the future. Some combinations, such as 222, 441,112, show meagre returns and are probably not worth following. I would also say that combinations that record relatively low strike rates are also worth ignoring because long losing runs are more likely. As a rule of thumb I wouldn’t want to bet on a combination that had a strike rate of less than 15 per cent. This is a rule that I would apply to all types of race and not just to handicaps.
37
PROFITABLE BETTING SYSTEMS Table 3.5: Non-juvenile handicaps Last Second Third Wins Total
% Win Profit/loss (£) Profit/loss (%)
run
Last run Last run
3
4
2
29
182
15.9
29.25
16.07
3
2
1
55
284
19.4
42.79
15.07
4
4
3
25
192
13.0
16.60
8.65
4
3
1
36
221
16.3
10.00
4.52
3
3
4
31
207
15.0
9.29
4.49
2
2
4
46
255
18.0
7.95
3.12
2
1
2
73
396
18.4
12.02
3.03
4
3
3
27
217
12.4
6.46
2.98
4
1
0
107
803
13.3
19.50
2.43
1
4
1
41
246
16.7
4.97
2.02
1
1
2
115
520
22.1
4.08
0.79
4
4
1
25
205
12.2
1.25
0.61
2
2
2
52
293
17.7
1.38
0.47
662
4,021
16.5
165.55
4.12
Total
Juvenile maiden races
Juvenile maiden races produce a profit of nearly 11 per cent and the strike rate for four of the five winning combinations is more than 15 per cent. The sample size for some combinations though is a bit on the small side and I would want to see a bit more action before placing my hard-earned cash on the form triples 340 and 330.
Table 3.6: Juvenile maiden races Last Second Third Wins Total Win % Profit/loss (£) Profit/loss (%) run
Last run Last run 3
3
0
16
67
23.9
22.00
3
4
0
12
55
21.8
10.56
19.20
2
2
2
22
70
31.4
8.06
11.51
38
32.84
FORM FIGURES Last
Second
run
Last run Last run
Third
Wins Total Win % Profit/loss (£) Profit/loss (%)
3
2
No run
32
87
36.8
7.11
8.18
0
3
0
24
189
12.7
2.50
1.32
106
468
24.7
50.23
10.73
Total
Juvenile non-maiden races This category records a terrific profit of just over 17 per cent, with some combinations recording very large returns. Some of the health warnings mentioned above apply, such as small sample sizes for some form triples and low strike rates, but there are plenty of combinations that return good profits, on large samples, and have a high strike rate. For instance, the figures 022 and 31 are of particular interest. Table 3.7: Juvenile non-maidens Last Second Third Wins Total
% Win Profit/loss (£) Profit/loss (%)
run
Last run Last run
2
1
3
24
91
26.4
85.42
2
0
2
26
97
26.8
37.13
38.27
0
2
2
20
107
18.7
37.81
35.34
93.87
0
1
1
25
168
14.9
53.96
32.12
3
2
1
15
69
21.7
14.23
20.62
4
1
No run
17
95
17.9
19.44
20.47
2
2
No run
10
52
19.2
9.79
18.83
4
0
2
11
77
14.3
13.01
16.89
2
4
0
21
98
21.4
15.60
15.92
2
2
2
18
69
26.1
9.27
13.43
2
0
4
18
92
19.6
12.25
13.32
0
2
1
16
152
10.5
17.75
11.68
3
1
No run
32
147
21.8
15.87
10.80
0
3
2
15
99
15.2
9.15
9.24
0
1
3
25
159
15.7
9.43
5.93
0
4
2
11
80
13.8
4.18
5.22
39
PROFITABLE BETTING SYSTEMS Last
Second
run
Last run Last run
Third
Wins Total
1
3
3
14
79
17.7
3.25
4.11
2
1
0
31
159
19.5
1.73
1.09
3
1
3
9
59
15.3
0.58
0.97
1
0
3
17
119
14.3
0.42
0.35
1
2
3
22
103
21.4
0.06
0.06
397
2,171
18.3
370.30
17.06
Total
% Win Profit/loss (£) Profit/loss (%)
Maiden races for three-year-olds and older horses Maiden races for three-year-olds and older horses are an interesting race category because races of this type often comprise a mixture of some horses that are of classic potential and other ones that are completely useless. From a form figure perspective they also provide rich pickings, showing a rate of return of nearly 18 per cent. Again, be aware of the health warnings that I have already mentioned and note combinations like 002, 222 and 330. They show decent profits on a high strike rate and are based on reasonable sample sizes. Table 3.8: Three-year-old plus maiden races Last Second Third Wins Total % Win Profit/loss (£) Profit/loss (%) run
Last run last run 0
4
3
16
76
21.1
52.37
68.90
3
2
2
31
82
37.8
43.80
53.41
4
0
3
17
79
21.5
28.25
35.76
4
2
2
19
62
30.6
20.76
33.48
0
2
4
20
82
24.4
18.28
22.29
3
2
3
22
62
35.5
12.29
19.82
2
2
4
19
51
37.3
9.81
19.24
0
4
0
51
411
12.4
60.88
14.81
0
0
2
51
287
17.8
37.37
13.02
2
2
2
41
107
38.3
9.38
8.77
3
3
0
28
117
23.9
9.92
8.48
40
FORM FIGURES Last
Second
run
Last run last run
Third
Wins
Total
% Win Profit/loss (£) Profit/loss (%)
3
0
2
21
86
24.4
4.94
5.75
4
0
2
14
59
23.7
2.91
4.93
0
3
4
16
96
16.7
2.27
2.37
0
2
2
35
114
30.7
1.09
0.95
401
1,771
22.6
314.3
17.75
Total
Nursery races
I also analysed the form figure combinations for nursery races but I couldn’t find any profits that were worth reporting.
Debutantes
You may wonder whether I considered the record of horses that were making their first racecourse appearance. I did and you should know that debutantes tend to record a win rate of between seven and eight per cent and you record a huge loss by blindly backing this type of animal. A study of the betting market, and a good knowledge of a horse’s breeding are probably the only ways to profit from previously unraced horses.
Conclusion
The simplest way to study form is to study form figure combinations. Most serious backers and racing professionals would sneer at such an approach, and for many years I personally felt that the method was naive. However, this probably explains why certain form figure combinations show a profit! If the serious backers are paying them little notice then it opens up the prospect of obtaining a bit of value.
41
PROFITABLE BETTING SYSTEMS
I would suggest that you back any one of the form figure combinations I have discussed, providing that the combination records a strike rate of at least 15 per cent. This will avoid long losing runs. You could be even more discerning than this and back only those form triples that have a strike rate of at least 20 per cent. However, please remember that the combinations only relate to turf and all-weather flat racing. They do not work for jump racing. This though still provides plenty of action.
On occasion you will find that a number of horses qualify for a bet in the same race. The figures I have produced above assume that you would back them all. You could be more selective and introduce some other filter. A good one might be to back the one with the shortest forecast price.
42
Chapter 4: The claiming jockey system The jockey plays a vital role in the outcome of a horserace. There is, in most races, a thin margin between winning and losing and the tactics employed by the jockey can be decisive.
In flat racing a jockey that sets too fast a pace, gets boxed-in, or who plays their hand too soon can easily turn a probable victory into a defeat. It is therefore important to be able to identify the more competent riders. The better riders make fewer mistakes and win races they should have lost by employing superior tactics.
The assessment of a jockey’s ability though can be a subjective process. Every punter has a view on the competence of particular riders. If you stand in a betting shop long enough you will hear fellow punters call a jockey all the names under the sun when they ride them a loser, but when they ride them a winner in the next race they are the hero. However, the systematic bettor doesn’t talk through their pocket. They adopt an evidenced-based approach, by calmly assessing the merits of each jockey against the available data.
The data most frequently used to assess a jockey’s race-riding ability are the number of winners ridden by each rider, and it is this data that is used to determine the annual jockeys’ championship. This sounds like a rationale approach. After all, the jockey who wins the most races, is clearly highly competent, otherwise trainers and owners wouldn’t give them rides.
PROFITABLE BETTING SYSTEMS A great number of punters base their selections around the top riders who are close to the top of the championship table. This isn’t a bad system because the top jockeys ride a high proportion of winners, particularly in the high-class races. However, the booking of a high profile jockey can often send a horse's odds plummeting. As a result you would waste plenty of your hard-earned cash if you blindly bet all the mounts of Dettori, Spencer and co. This raises the question of whether it is possible to develop a betting strategy based solely around jockeys?
Claiming jockeys
I have found that one of the more profitable systems is to look beyond those jockeys that are competing for the title of champion jockey, and instead concentrate on trying to identify the new, young, up-and-coming apprentice jockeys. The punter who is able to spot the rising stars of the future will tend to strike value bets because young, low profile apprentices do not attract other punters’ money.
The additional advantage of betting on jockeys that are just starting their careers is that their mounts benefit from a weight allowance to compensate for their jockey’s lack of experience. These weight claims can be used when the apprentice jockey is competing against established professionals.
The level of the allowance varies according to the number of winners ridden by the jockey. An apprentice that has ridden fewer than 20 winners will earn a claim of seven pounds. A jockey that has ridden more than 20 winners but less than 50 will receive an allowance of five pounds. The jockey that has ridden between 51 and 95 winners will claim an allowance of three pounds.
44
THE CLAIMING JOCKEY SYSTEM Once a jockey has ridden more than 95 winners they lose their claim. Typically this is when most young jockeys struggle because they are then competing for rides against more experienced riders, but don’t have the advantage of a weight allowance. As a result trainers don’t have an incentive to give them rides, and prefer to employ the services of the more experienced jockeys.
Betting claiming jockeys
At first glance, betting horses ridden by claiming jockeys looks like a one-way ticket to the poor house. The bare statistics reveal that they have a dreadful win record (see Table 4.1), and huge losses would be recorded by backing their mounts. The most disastrous strategy would be to back horses ridden by jockeys receiving the maximum seven-pound allowance. They record, on average, just 6 wins for every 100 rides and you would make a loss of around 40 per cent. In other words for every pound you handed over to your bookmaker you would receive just 60 pence in return. Ouch! Table 4.1: Record of claiming jockeys by weight allowance, all races. Claim (lbs) Winners Total rides % Win Profit/loss (£) Profit/loss (%) 0
27,152
273,773
9.9
-86138.78
-31.46
3
2,062
24,915
8.3
-7238.88
-29.05
5
2,151
27,883
7.7
-8852.57
-31.75
7
1,474
23,029
6.4
-9201.91
-39.96
The three-pound and five-pound claiming jockeys do a bit better. Those jockeys claiming five pound have a win rate of around 8 per cent and make a loss similar to that recorded by established jockeys (i.e. those not in receipt of a claim) The three-pound claimers are even more interesting. They are the most experienced claiming jockeys and they return to the winners’ enclosure
45
PROFITABLE BETTING SYSTEMS eight times for every hundred rides and make a loss of just over 29 per cent. This is slightly better than for the non-claiming group of jockeys.
The figures look slightly better when one considers only handicap races (see Table 4.2).
In handicap races, seven-pound claimers still have an appalling record, which suggests that the weight allowance that they receive is not sufficient to compensate for their lack of race-riding experience. The three and five pound claimers have a win record more comparable to the established jockeys and, more importantly, they record the lowest level of loss. The three-pound claimers have the best record. They make a loss of just over 24 per cent. This is more than four percentage points lower than for the non-claiming jockeys and almost 12 percentage points lower than the seven pound claimers. Table 4.2: Record of flat claiming jockeys by weight allowance, handicap races. Claim (lb) Winners Total % Win Profit/loss (£) Profit/loss (%) 0
11,584
132,826
8.7
-37686.25
-28.37
3
1,216
15,020
8.1
-3607.15
-24.02
5
1,433
18,294
7.8
-4792.04
-26.19
7
926
14,431
6.4
-5174.99
-35.86
Thus far, the evidence is clear that claiming jockeys do best in handicap races, and that the three and five pound claiming jockeys record the lowest level of loss. However, this information is only of academic interest. There are no profits to be made from exploiting this knowledge. The situation though can be radically transformed if a more selective approach is taken.
The most efficient strategy, in theory at least, would be to follow those claiming jockeys that one judges to be highly competent and to back their
46
THE CLAIMING JOCKEY SYSTEM mounts before other punters catch onto their ability. Unfortunately this is easier said than done.
I have personally never ridden a racehorse and I accept that as a result I'm probably not the best judge of a jockey’s ability. However, to make a profit from backing claiming jockeys one doesn’t need to be an expert. The trainer will do all the work for you, because the more astute trainers are particularly adept at employing the most competent claiming jockeys. They employ them regularly in order to exploit their valuable weight allowance. The sensible strategy is therefore to analyse those trainers that record a high proportion of winners when employing claiming jockeys.
I have analysed the record of all flat race trainers that employed the services of apprentice jockeys at least 30 times during the last few turf and all-weather racing seasons. I excluded all those trainers that had fewer than 30 runners ridden by apprentices in order to generate more statistically reliable findings. I also only analysed the record of trainers that employed three or five pound claimers. This appeared to be a sensible strategy on account of the fact that seven pound claiming jockeys generally have an appalling win record.
The results of my analyses are presented in Table 4.3, and are truly staggering. The data suggests that you would earn a profit of over 50 per cent by simply backing the runners of the selected trainers when they employ the services of 3 and 5 pound claiming jockeys. Table 4.3: Flat trainers record employing 3lb and 5lb claiming jockeys, turf and all-weather flat race seasons. Trainer Winners Total % Win Profit/loss Profit/loss (%) M D I Usher C G Cox
17
65
26.2
100.95
155.31
6
30
20.0
37.90
126.33
47
PROFITABLE BETTING SYSTEMS Trainer
Winners
Total
% Win Profit/loss Profit/loss (%)
Dr J R J Naylor
7
37
18.9
50.50
136.49
Mrs P Townsley
7
39
17.9
58.00
148.72
R Johnson
5
30
16.7
16.00
53.33
M S Saunders
13
84
15.5
89.50
106.55
L M Cumani
15
101
14.9
1.96
1.94
M Wigham
5
35
14.3
4.50
12.86
D W Barker
13
93
14.0
81.50
87.63
D Nicholls
10
73
13.7
0.70
0.96
S R Bowring
12
89
13.5
19.80
22.25
J R Best
15
113
13.3
73.00
64.60
Ian Williams
9
71
12.7
41.00
57.75
K A Ryan
9
71
12.7
9.50
13.38
B A Pearce
5
40
12.5
16.00
40.00
W R Muir
4
32
12.5
10.50
32.81
Mrs C A Dunnett
20
161
12.4
153.83
95.55
B Ellison
24
202
11.9
5.20
2.57
7
63
11.1
27.25
43.25
R M Beckett
12
109
11.0
78.30
71.83
Miss J Feilden
16
150
10.7
105.17
70.11
E J Alston
D R C Elsworth
8
76
10.5
77.75
102.30
M H Tompkins
11
105
10.5
29.50
28.10
G L Moore
10
99
10.1
22.63
22.86
J A R Toller
5
52
9.6
6.50
12.50
J R Jenkins
4
43
9.3
15.50
36.05
R Bastiman
4
45
8.9
22.50
50.00
C E Brittain
4
46
8.7
13.50
29.35
Mrs H Sweeting
5
59
8.5
13.00
22.03
11
157
7.0
38.50
24.52
P Howling
2
30
6.7
44.00
146.67
G A Ham
2
31
6.5
2.50
8.06
J M Bradley
J R Weymes
5
98
5.1
0.30
0.31
D A Nolan
1
41
2.4
26.00
63.41
313
2,621
11.9
1319.75
50.35
Total
48
THE CLAIMING JOCKEY SYSTEM However, one needs to be cautious when one sees such profitable returns. You will note that the data in Table 4.3 is ordered by trainers’ strike rate. A number of the trainers have very low strike rates. For instance, Nolan (2.4 per cent) and Weymes (5.1 per cent) had 139 runners between them but recorded just 6 winners. The winners though were returned at very high odds, and this accounted for the profits. I wouldn’t want to be backing the runners from such stables, and indeed I wouldn’t want to rely on trainers that recorded a strike rate of less than 10 per cent. Employing this rule gives a more select list of trainers to follow and actually increases the rate of return from 50 to 56 per cent. In Table 4.4 I have repeated the above analysis for jump races, and again there is a group of trainers that appear to be worth following when employing claiming jockeys. Table 4.4: Jump trainers’ record employing 3lb and 5lb claiming jockeys. Trainer Winners Total %Win Profit/loss(£) Profit/loss(%) BOWE J
27
83
32.5
1.91
2.30
DENNIS C P
11
35
31.4
9.25
26.43
O`KEEFFE J
11
35
31.4
0.96
2.75
BAILEY MRS CAROLINE
37
129
28.7
42.44
32.90
DARTNALL V
13
47
27.7
7.57
16.10
HUTSBY F A
10
39
25.6
10.36
26.56
PIKE S
15
60
25.0
12.42
20.70
EGERTON C R
148
606
24.4
42.06
6.94
FANSHAWE J R
39
160
24.4
37.02
23.14
WARNER W J
11
46
23.9
14.52
31.57
BORRADAILE MRS L
16
67
23.9
12.83
19.15
O`NEILL J
10
42
23.8
14.34
34.15
CHANCE N
22
95
23.2
14.98
15.77
DOUMEN F
52
228
22.8
51.88
22.75
DENNIS W W RICHARDS N G
10
44
22.7
6.72
15.28
302
1352
22.3
11.89
0.88
49
PROFITABLE BETTING SYSTEMS Trainer
Winners Total %Win Profit/loss(£) Profit/loss(%)
TATE T P
65
294
22.1
77.19
26.25
BALDING A M
19
87
21.8
28.50
32.76
650
3003
21.6
194.62
6.48
FITZGERALD J
31
145
21.4
58.57
40.39
TINKLER C
48
233
20.6
53.89
23.13
TWISTON-DAVIES NA
19
93
20.4
2.68
2.88
WAREHAM G
11
54
20.4
37.60
69.63
WORMALL MISS J
20
102
19.6
30.70
30.09
GROUCOTT J
14
72
19.4
31.16
43.28
O`BRIEN M J P
95
500
19.0
0.24
0.05
JONES T L
12
65
18.5
5.58
8.58
HAZELL M G
10
56
17.9
20.00
35.71
JONES T THOMSON
17
96
17.7
16.25
16.93
HAMILTON MRS A
27
155
17.4
9.90
6.39
MEADE M
10
58
17.2
64.33
110.92
HERRIES LADY
33
192
17.2
10.28
5.35
GREATHEAD T R
16
99
16.2
95.19
96.15
BERRY A
12
76
15.8
18.08
23.79
FLOOK S
21
134
15.7
1.10
0.82
WILLIAMS EVAN
142
910
15.6
138.35
15.20
WILLIAMS NICK
46
299
15.4
62.96
21.06
MURPHY C A
39
254
15.4
3.38
1.33
RICH P M
23
151
15.2
22.13
14.65
MANN C J
258
1703
15.1
66.08
3.88
ELSWORTH D R C
42
278
15.1
49.65
17.86
HOGARTH H P
22
150
14.7
13.99
9.33
243
1683
14.4
10.86
0.65
SHERWOOD S E H
56
397
14.1
66.18
16.67
DUN J M
11
80
13.8
32.38
40.47
FITZGERALD J G
55
403
13.6
14.50
3.60
EDWARDS MISS S
11
81
13.6
6.41
7.91
WEST MISS S
33
245
13.5
161.60
65.96
JOHNSON MRS A E
15
113
13.3
18.00
15.93
CONDON K J
12
92
13.0
6.03
6.55
HENDERSON N J
GEORGE T R
50
THE CLAIMING JOCKEY SYSTEM Trainer
Winners Total %Win Profit/loss(£) Profit/loss(%)
HUBBARD G A
16
124
12.9
6.88
5.54
FORSTER D M
20
156
12.8
43.63
27.96
ADAM J R
24
188
12.8
12.68
6.75
RIMELL M G
19
149
12.8
2.70
1.82
BRADSTOCK M
49
391
12.5
38.62
9.88
BANKS M C
11
91
12.1
33.35
36.65
MCINNES SKINNER MRS T J TAYLOR MRS L C
10
84
11.9
47.50
56.55
54
472
11.4
69.04
14.63
SMITH MISS S
13
115
11.3
19.50
16.96
JONES P J
11
98
11.2
25.50
26.02
BRAMALL MRS S
11
98
11.2
6.21
6.34
RYALL B J M
40
371
10.8
18.81
5.07
POGSON C T
22
211
10.4
148.63
70.44
BARNETT G
10
96
10.4
70.50
73.44
NOCK MRS SUSAN
21
206
10.2
29.96
14.54
RUSSELL MISS LUCINDA
10
99
10.1
45.50
45.96
BISHOP K
87
862
10.1
101.38
11.76
WILLIAMS MRS S D
45
466
9.7
82.38
17.68
ROBESON MRS P
32
342
9.4
94.85
27.73
HAMER MRS D A
32
347
9.2
77.21
22.25
SADIK A
12
140
8.6
11.00
7.86
ROBERTS C
21
249
8.4
94.50
37.95
NEWCOMBE A G
16
190
8.4
36.13
19.02
DUKE B W
16
196
8.2
124.33
63.44
KIRBY F
15
187
8.0
5.38
2.87
PAYNE J R
10
126
7.9
11.50
9.13
SMITH R J
18
229
7.9
15.81
6.90
DODDS J P
11
140
7.9
22.55
16.11
BILLINGE J N R
16
208
7.7
9.19
4.42
COLTHERD W S
16
211
7.6
7.75
3.67
HALDANE J S
14
194
7.2
47.50
24.48
CLUTTERBUCK K F
15
219
6.8
9.02
4.12
CROW A M
10
152
6.6
3.13
2.06
CHEESBROUGH P
11
171
6.4
4.67
2.73
51
PROFITABLE BETTING SYSTEMS
I did reanalyse the results for flat and jump races to look at handicap and nonhandicaps. This didn’t seem to make much difference. The results also looked less reliable because they were based on smaller samples. Conclusion
It is possible to develop profitable betting systems based around three and five pound claiming jockeys, provided that they are riding for trainers that recognise talent when they see it.
52
Chapter 5: A classy system I have purchased and developed a number of betting systems that have used a measure of a horse’s class as a variable. Indeed, every book on form study ever written has claimed it to be an essential winner-finding factor. What though is class?
In answer to this question it first important to get the terminology clear. For me the term class is simply another word for ability and I use the terms interchangeably. Class (or ability) refers to a horse’s physical ability to run faster than another, over a given set of race conditions. These conditions are defined by the race distance, the going, the pace of the race, the jockey’s ability, the fitness level of the horse, the configuration of the track, etc.
In theory, if it was possible to hold all external conditions constant, and each horse in the race was ideally suited to them, then the class horse in the race would be obvious. The most able horse would win every time. However, this is theory and the reality is more complex. The fact is that the class horse does not win all the time, and this is because the class horse may be unsuited by the conditions. In other words if a horse possesses a high level of physical ability but is unfit, is ridden by a useless jockey, encounters ground that it hates, or races over a distance that is either too short or too far then it will under-perform. All these factors can therefore mask a horse’s true level of ability. The expert form analyst needs to identify a horse’s true level of ability and work out the race conditions that will allow it to run to that level. This is the whole basis for picking winners, and it is not an easy task.
PROFITABLE BETTING SYSTEMS Most betting systems based around a horse’s class assume that it is possible to quantify ability in terms of a rating. Herein lies a significant problem for the study of class for, while everyone accepts that it exists and recognises its importance, there is no agreement about how to measure it.
Official Handicap ratings
The official handicap ratings are not a bad place to start when trying to build a class rating into a betting system. Indeed, handicap ratings are probably the most common form of ability rating. They are steeped in tradition, with the first principles of handicapping being established as long ago as 1855 by Admiral Rous, when he became the first official handicapper. For those who are not familiar with the handicapping process I will briefly explain how and why they are constructed.
Handicap ratings place every horse on an ascending scale range. In flat racing the scale ranges from 0 to 140 and in jump racing it ranges from 0 to 170. In theory at least, the rating should reflect the horse’s level of ability, with the more able horses having a higher rating. The official ratings compiled by the British Horseracing Authority (BHA) are used to allocate weights in handicap races and to assign horses to particular types of race.
The official ratings are expressed in pounds, with one point on the scale being equal to 1 pound in weight. In this way the handicapper can assign a horse with a weight when they run in a handicap race. For example a horse with a rating of 100 would carry 10 pounds more in weight than a horse running off a rating of 90 points. The aim of the handicapper in assigning these weights is to give each horse an equal chance of victory. Later in this chapter we will consider whether this theory works in practice.
54
A CLASSY SYSTEM The important thing to note is that the ratings themselves are not objective, although they do follow certain principles. They are the subjective opinion of the handicapper. Indeed there are various forms of handicap ratings, with a range of different organisations providing them. For example, the BHA provide the official ratings, but most punters would be more familiar with the Racing Post Rating (RPR). This is the Racing Post’s own private handicap ratings service. However, while there are different ratings services, all handicap ratings are compiled in a similar way. Thus while I said in Chapter two that you should avoid subjective variables when developing systems, provided that you use a consistent source of information (i.e. don’t mix and match data sources) the class rating you use should be fairly objective and reliable over time.
A few years ago I developed a betting system based around British Horseracing Board ratings (BHB), as official handicap ratings were called at the time. This system has continued to perform well.
When I originally developed the system I analysed the results of non-handicap flat races run in Great Britain over a 12-year period. Why? Well nonhandicaps seemed to be fertile ground for handicap ratings. Following the principles that I set out in Chapter two, I split my data into two parts. I used data from 10 years to test various hypotheses about BHB ratings. I then developed these into a system that I tried to test on the two remaining seasons in my sample. I excluded those races where one or more of the runners did not have a BHB rating. Thus I was only looking at those races where all the runners had exposed form. This seemed sensible, on the assumption that it was in these circumstances that poundage handicaps would perform best because the handicapper had had a good look at how the horses had performed in the past.
55
PROFITABLE BETTING SYSTEMS Table 5.1 shows that if you had bet on the horse top-rated on BHB ratings you would have made 2,150 selections and scored 532 wins (24.7per cent). You would however have made a loss of £252.89 or 12 per cent. Not a great result I have to admit but a high proportion of winners were concentrated in the top three rankings (57 per cent), and so the method is a good way of narrowing down the field to a few live contenders. It also should be noted that in a number of races there was more than one top-rated selection, which biases the results a little.
These results though are based on raw BHB ratings. They have not been adjusted for weight carried or the weight-for-age scale. However, surprisingly, re-running the analysis on adjusted figures (identical to those published in the Racing Post ratings feature) does not improve the situation. The strike rate on the top-rated selection under this method is only 20 per cent and losses are considerable (-21%). This is all a bit illogical on first reading, and it certainly goes against the conventional wisdom that ratings need to be adjusted for weight carried. However, on reflection these results are explicable.
It is a statistical fact that the higher a horse’s handicap weight the greater its chances of success. This applies to handicaps and non-handicap races alike. In non-handicaps, horses race off different weights if they receive allowances (for age and/or sex) or penalties because they have either recently won a race, or they won a race above a certain value. The penalties reflect the fact that the horse is superior to its rivals. The penalties though do not appear to be sufficient to reduce the horse’s likelihood of success.
I will now make the assumption that racehorse trainers are not stupid. They will place their horses in races that incur them an advantage. For example, if a trainer enters a horse in a race with a seven-pound penalty he probably thinks
56
A CLASSY SYSTEM that the horse has at least that in hand and this negates the extra burden. For these reasons (and probably a few more) extra weight is a positive factor. It reflects a horse’s level of ability. The fact that unadjusted BHB ratings outperform adjusted ratings in non-handicaps suggests that weight differences reflect form factors not necessarily accounted for in the official ratings. One should therefore not attempt to control for weight. In fact one should actually see extra weight as a positive and up-rate ratings accordingly.
Following this approach I recalculated BHB ratings by adding weight carried (in lbs). For example, a horse rated 100 and carrying 9 stone (i.e. 126 pounds) would be rated 226 (100 + 126 pounds = 226). This method (see Table 5.1) actually produced slightly better results than using the raw rating alone. The strike rate increased to 25.4 per cent and the loss on turnover went down to 6.6 per cent. I didn’t factor in weight-for-age and so in effect younger horses taking on their elders, with the same raw BHB mark, would receive a lower rating. My analysis though showed that WFA didn’t really make a difference to results. In fact they were slightly better without this adjustment. Table 5.1: Adjusted and unadjusted BHB ratings Type of rating
Wins
Rnrs
%Win Profit/loss(£) Profit/loss(%)
Raw BHB rating
532
2,150
24.7
-251.5
Raw BHB rating+weight
473
1,865
25.4
-122.51
-11.7 -6.6
Raw BHB rating+weight+WFA
484
2,380
20.3
-494.97
-20.8
Raw BHB rating+weight+sex allow
491
1,884
26.1
-71.28
-3.8
The mares’ or fillies’ sex allowance though did seem to make a difference. Adding this back to the weight + raw BHB rating produced the most successful form of ability rating, with top-rated selections winning 26.1 per cent of all their starts. For example, a mare or filly with a raw BHB rating of 100 and carrying 126 pounds would earn a rating of 226. Adding back the five
57
PROFITABLE BETTING SYSTEMS points (i.e. five pounds) for her sex allowance gives the mare a final rating of 231. The success of this rating formula was the basis for a profitable system.
Using the ratings in the right circumstances
One way of turning the ratings method into a profitable system would be to apply it in conditions which favour highly weighted horses, namely races run on flat, downhill or tight tracks and on ground no softer than good. In these conditions it is much easier for horses to carry high handicap weights. Nick Mordin, for example, showed several years ago that top-weighted horses perform well on tight tracks because the horses can conserve their strength whilst the field has to slow down to negotiate the bends. Flat and downhill tracks also make it easier for horses to carry weight, compared to courses with stiff up-hill finishes (Table 5.2). In addition it is far easier to carry a high handicap weight on firm ground than on a soft surface (see Table 5.3). The effect of extra weight also seems to make more of a difference over sprint distances when the margin between winning and losing is slight, and could be the affected by weight carried (see Table 5.4).
Table 5.2: Top rated by type of track Track type
Winners
Runners
%Win
Stiff
52
239
21.8
Tight
200
705
28.4
Galloping
132
542
24.4
Downhill
28
112
25.0
Non-descript
79
286
27.6
Stiff- Beverley ,Carlise, Ascot, Leicester, Newcastle, Sandown Tight- Catterick, Musselburgh, Epsom, Folkestone, Kempton, Lingfield, Southwell, Thirsk, Wolverhampton, Windsor, Chester Galloping- Ayr, Doncaster, Haydock, Nottingham, Redcar, Salisbury, Yarmouth, York, Newmarket, Newbury Downhill - Brighton, Chepstow
58
A CLASSY SYSTEM
Table 5.3: Top rated by type of surface Surface Winners Good or firmer
%Win
1,234
25.9
59
273
21.6
112
372
30.1
Good-to-soft or softer All-weather
Runners
319
Table 5.4: Top rated by distance Distance group Winners
Runners
%Win
5f to under 6f
120
537
6f to under 8f
153
561
27.3
8f to under 12f
179
659
27.2
39
127
30.7
12f plus
22.3
The optimum conditions for the rating formula would therefore appear to be in races run over distances of six furlongs or more, run on good ground or ground firmer than good (or all-weather surfaces), and run over ‘easy’ tracks. In addition, if the horse hasn’t run for more than 28 days then its rating is history, and if it didn’t finish within the first six on that run then it is probably out of form. Excluding horses with these characteristics further increases the strike rate for the top-rated selections.
The rating formula could be also be improved by considering those top-rated selections that have a significant advantage over their rivals. Indeed without applying any of the above filters we can hit nearly 30 per cent winners by simply selecting only those horses that have at least a pound in hand over the next highest rated runner (see Table 5.5). Table 5.5: Rating advantage of top rated selections Adv (lbs) Wins Runners %Win No adv 1+
37
183
20.2
564
1,901
29.7
59
PROFITABLE BETTING SYSTEMS Adv (lbs)
Wins
Runners
%Win
2+
508
1,677
30.3
3+
452
1,472
30.7
4+
408
1,286
31.7
5+
361
1,131
31.9
6+
326
983
33.2
7+
298
872
34.2
8+
280
769
36.4
9+
265
673
39.4
10+
247
598
41.3
Making a small profit
If we had applied the rating method in optimum conditions, to races in the development sample, we would have placed a total of 495 wagers and hit 174 winners (35.2 per cent). Some years would have been better than others, but overall we would have earned ourselves a small profit of over 7 per cent on turnover at SP odds.
Producing profitable results from past data though is easy. We can all be wise after the race has been run! In the above analysis I have added various filters to the results to improve the results. However, as described in Chapter two, the acid test for any system is whether it produces results on ‘unseen’ races. To validate the system I therefore applied it to ‘unseen’ races run in two other seasons. In these years we would have placed 199 bets and would have had been paid out 71 times (35.7 per cent), for a profit of 9.1 per cent on turnover. If we had been prepared to put in a bit more work to obtain the best odds on each selection then we could no doubt have improved the rate of return.
60
A CLASSY SYSTEM
Conclusion In summary the system rules are: x
Consider only non-handicap flat races.
x
Race distance must be six furlongs plus.
x
Going is reported as good or firmer.
x
Do NOT apply the system to races run at stiff tracks like Beverley, Carlisle, Ascot, Leicester, Newcastle, and Sandown.
x
Bet the horse with the highest system rating (i.e. official rating + weight + sex allowance).
x
Must be the clear top-rated selection (i.e. ignore horse’s that are joint top rated).
x
Last ran within 28 days and finished in the first six.
My recent research into this system shows that it performs reasonably well and if one applies the ratings in the right circumstances then a small profit should be made. The amount of action is limited to around 100 bets per season. It isn’t a get rich quick scheme (few honest systems are) but it continues to turn in a few quid in most years.
61
Chapter 6: betting market systems You can spend a lot of time studying the form. I know of at least one leading racing tipster who practically leads a nocturnal existence by spending all night analysing the form of the runners for the following days racing. There are no doubt countless other professional and semi-professional punters who follow the same lifestyle in order to make their racing pay. I have certainly been part of this group from time to time, but in recent years, with the arrival of two young children, sleep has become more of a priority! I have therefore looked at shortcutting the selection process by developing systems that use the betting market as a guide to winners.
The betting market is a good short cut because it basically reflects the views of the thousands of punters and bookmakers who have bothered to study the form. As a general guide the betting market is not too far off the mark, and the odds on offer closely match the actual statistical probabilities of a horse making it into the winners’ enclosure (see Figure 6.1).
BETTING MARKET SYSTEMS Figure 6.1: Betting market odds expressed as a percentage probability of a win and compared to actual percentage wins
100.0
90.0
80.0
Actual % probability of win
70.0
60.0
50.0
40.0
30.0
20.0
10.0
0.0 0.0
10.0
20.0
30.0
40.0
50.0
60.0
70.0
80.0
90.0
100.0
Market odds(%probabilityof win)
Figure 6.1 pays for careful consideration. Basically it shows the relationship between the betting markets’ assessment of a horse’s chances and the actual probability of the horse winning.
The Starting Price (SP) betting odds have been converted into statistical probabilities. To give you an example, an even money chance (1/1) is
63
PROFITABLE BETTING SYSTEMS assumed by the market to have a 50 per cent probability of winning. A 2 to 1 shot would be said to have a 33 per cent chance of success because the bookmakers are basically saying that in every three races a 2 to 1 shot will win one race and lose two.
The betting market probabilities have then been compared to the actual probability of a win. These ‘actual probabilities of a win’ are simply the proportion of winners recorded in all flat races run in Great Britain and Ireland over the last decade for horses at different betting odds. This gives a sample of over 700,000 runners, which allows for robust comparisons to be made.
The first interesting thing to note about Figure 6.1 is that the solid line represents a perfect relationship between the betting odds and the actual probabilities. This would reflect the ‘fair odds’ i.e. the betting odds represent the true probability of a horse winning. However the dots, which are the real observations, are mostly below this line. This means that the betting market odds overstate the true probability of success. For example, even money shots, in a fair market, should win 50 per cent of all their races. In actual fact even money shots win only 46 per cent of their races. The difference shows that the bookmakers build into their odds a margin, which means that the odds are stacked in their favour. This to me illustrates just how difficult it is to make a profit betting on horses. In my view you will only win in the long run by getting value, and by this I mean that the true odds of a horse winning needs to be higher than the odds offered by the bookies. In other words you need to be striking bets at even money on horses that have more than a 50 per cent chance of winning.
64
BETTING MARKET SYSTEMS
The favourite-longshot bias
The other interesting thing about Figure 6.1 is that, while generally the market odds offer poor value in relation to the actual odds of winning, there is value to be had in backing short priced odds-on favourites. This is what is called the favourite-long shot bias.
Basically very short priced favourites are more likely to win than the odds suggested by the market, while longer priced runners are much less likely to win compared to their betting odds. This situation seems to come about because there is a price below which most punters are not prepared to bet. To give an example, a horse originally priced up as a 1 to 7 shot would attract very little money in the market. Most punters will simply ignore favourites at this price and will focus on betting on the longer priced runners in the hope that the favourite will get beat. This can create a bit of value around short priced favourites because their odds tend to lengthen because bookies are desperate to attract bets on them in order to balance their books.
The favourite-longshot bias can be exploited by simply betting on all starting price favourites priced at less than 1 to 4. However, this wouldn’t make you rich. You would make a profit of around 2 to 3 per cent on turnover, but this is much better than the 34 per cent loss you would have made had you simply backed every horse priced at odds greater than 4 to 1 against!
65
PROFITABLE BETTING SYSTEMS The dead eight system
There is another slight edge you can gain over the market by applying what is called the dead eight system. This is quite a neat system. It exploits a blind spot in the market concerning each-way bets in eight runner races.
As you probably know you can place an each-way bet in a seven-runner race and if your horse finishes third the bookmakers wouldn’t pay out. The horse would need to finish first or second in a seven-runner race in order to qualify as a successful each-way bet. However, in an eight runner race the rules change and you would be paid out on a horse that finished third.
This gives you a statistical edge because you now have three chances to collect on your each-way bet compared to just two in a race with between five and seven runners. In terms of the percentages this means that you have a 38 per cent chance of collecting on an each way bet in an eight runner race (3 possible places/8 runners = 38 per cent) compared to just 29 per cent in a seven runner race (2 possible places/7 runners = 29 per cent). In Table 6.1 I have detailed the probabilities of landing a successful each-way bet in races with between 5 and 30 runners. Table 6.1: Probabilities of achieving a pay-out on an each-way bet Runners No of EW places % Prob of placing 5
2
6
2
40.0 33.3
7
2
28.6
8
3
37.5
9
3
33.3
10
3
30.0
11
3
27.3
12
3
25.0
66
BETTING MARKET SYSTEMS Runners No of EW places % Prob of placing 13
3
23.1
14
3
21.4
15
3
20.0
16
4
25.0
17
4
23.5
18
4
22.2
19
4
21.1
20
4
20.0
21
4
19.0
22
4
18.2
23
4
17.4
24
4
16.7
25
4
16.0
26
4
15.4
27
4
14.8
28
4
14.3
29
4
13.8
30
4
13.3
Naturally some of the statistical biases in each-way betting are reflected in the betting odds but you can nevertheless back the second favourite each-way in an eight runner race when the favourite is odds-on and make a small profit of between 7 and 8 per cent on turnover. For instance, over the last few years you would have made 505 bets on this system and landed 328 successful each-way bets (a strike rate of over 65 per cent) returning a level stakes profit of £37, or a rate of return of 7.4 per cent.
The astute of you would also have noticed that each-way betting on five runner races also has a high probability of success. I can tell you that betting the second favourite each-way in these races when the favourite is odds-on also yields a small profit of around three per cent on turnover.
67
PROFITABLE BETTING SYSTEMS
Implementing betting market systems
Systems based around the betting market yield modest returns but they are very consistent and you have a high probability of success that psychologically is very uplifting, especially when you are betting to large stakes.
The downside is that you need to be watching the betting market very closely in order to strike the right bet at the right time. The dead eight system described above depends on backing the second favourite. This though is easier said than done because you don’t know which horse is definitely going to start the race as the second favourite. You therefore have to watch the market carefully and place your bet at the last possible moment before the off, to be sure that you are on the second favourite. Obviously in some races it is pretty clear which horse will be the second favourite but when you have an odds-on favourite in the race you need to keep a close eye on which horse will go off second best in the market. This is because the odds of the other runners can converge. In the event of joint or co-second favourites you should give the race a miss as you probably will not obtain odds high enough to justify each-way bets on multiple selections.
As well as watching out for the second favourite in the market you also need to make sure that there is an odds-on favourite in the race. This needs to be watched for closely if a favourite is trading in and around odds of 10 to 11 or even money. Similarly if you are trying to exploit the odds-on favouritelongshot bias you need to be looking to make sure that the favourite is to go off at odds of less than 1 to 4. I suggest that in order to maintain this level of
68
BETTING MARKET SYSTEMS concentration you have a pot of coffee on the go in order to keep the blood flowing to the brain.
Let us now turn to one of my favourite systems (pardon the pun), which I call the beaten favourites system. The beaten favourites system
In the average racing season if you backed every favourite you would back roughly one winner for every three bets that you made. This is an impressive strike rate and one that none of the tipsters in the national and sporting press could not match. However, you would still end up flat broke. For every pound that you staked on the favourite you would be lucky to average a return of ninety pence.
The favourite, though, is so tempting to back. It is often the horse with the best form, the best jockey, and the best trainer. It more or less jumps out at you when you open your racing paper, or review the runners in the parade ring. It looks fit, its gallop reports are scintillating, it has the highest official rating, and it has a set of cracking speed figures. A quick glance at your paper tells you that every racing pundit that you respect has made it his or her nap of the day. It can’t get beat! Sadly it does and becomes yet another beaten favourite.
However, in recent years I have come to love beaten favourites. When I view the runners and riders for the day’s racing in my newspaper and I see the initials BF (the standard term which denotes a horse was beaten as favourite on its most recent outing) after the horse’s name I immediately become interested in backing that horse. This is because I know that beaten favourites
69
PROFITABLE BETTING SYSTEMS have a good strike rate and a profit can be made if they are bet in certain circumstances.
The rationale for backing beaten favourites
The rationale for backing beaten favourites is obvious. For a horse to be sent off the market leader it has to have attracted a lot of money in the betting ring, and this essentially means that it was considered by other punters to be a good thing the last time it ran. It probably demonstrated a previous good performance, had the right connections and had proven fitness. There may have also been professional money for the horse. This could have come from the horse’s connections that may have been aware that the animal was particularly fit, and that it had been working well with the stable’s best animals. The fact that the horse started favourite and was beaten may have simply been due to bad luck in running (i.e. the horse was badly hampered when making a forward move in the race). Alternatively the horse may have run below expectations because it incurred some other misfortune such as losing a shoe during the race, or because it had contracted an illness that did not come to the attention of the trainer until later. The horse may also have been beaten because the distance of the race or the going was unsuitable. This again might only have been apparent in retrospect. The fundamental fact though is that the horse was intrinsically good enough to warrant market support and the fact that it was beaten may have been down to unforeseen factors that may very well not apply the next time the horse races.
70
BETTING MARKET SYSTEMS The record of beaten favourites
The rationale for backing beaten favourites is supported by the data in Table 6.2. This shows that beaten favourites generally have a decent win record the next time they race.
In this large sample of flat races, a total of 23,183 previously beaten favourites went to post and 3,642 of them returned to the winner’s enclosure. This represents a strike rate of almost 16 per cent. This compares to a rate of around 9 per cent recorded by all other runners. Table 6.2: Record of beaten favourites on their next start Winners Total % Win All other runners
27,004
295,238
9.1
Beaten favourite
3,642
23,186
15.7
30,646
318,424
9.6
Total
The record of beaten favourites is better still when one considers only those beaten favourites that are once again strongly supported in the betting market.
The data presented in Table 6.3 shows that if you bet every horse that was a beaten favourite on its last start and was starting favourite again then you would back one winner for every three selections. This would return you a loss of just three per cent. This is encouraging because Table 6.3 also shows that if you simply bet every favourite, regardless of whether it was a beaten favourite or not, you would make a loss of 7.6 per cent. Furthermore, if you bet every favourite that was not a beaten favourite on its previous start then you would make a loss of 8.4 per cent.
71
PROFITABLE BETTING SYSTEMS Table 6.3: Record of beaten and non-beaten favourites when made favourite again on next start Winners Total % Win Profit/loss (£) Profit/loss (%) Non-beaten favourite
8,784
29,192
30.1
-2454.29
-8.41
Beaten favourite
1,735
5,238
33.1
-162.79
-3.11
10,519
34,430
30.6
-2617.09
-7.60
Total
Clearly compared to other types of jollies those market leaders that were beaten favourites on their previous start are the ones to bet on, yet you would still make a small loss if backing them all religiously. How can this be turned into a profit?
Races restricted to two-year-olds.
The betting market has traditionally been a better guide in races restricted to two-year-olds. In these races the market favourite tends to have a relatively good record because the un-informed backer leaves these races alone. The odds on offer therefore reflect the amount of ‘smart’ money placed on each horse. This smart money generally comes from traders with “insider” knowledge, like the horse’s trainer, owner or others connected to the horse’s stable. This at least is the theory. Whatever the true reasons, it is possible to make a small profit if you bet only those beaten favourites that run in races restricted to two-year-olds, and are starting favourite again.
The data in Table 6.4 shows that if you confine your bets to two-year-old races and back every favourite that was previously a beaten favourite you could expect a small profit of around 3 per cent. You would also be returning to the pay out window on a regular basis because two-year-old beaten favourites record a strike rate of over 40 per cent! This is superior to the record of previously beaten favourites aged three years and upwards.
72
BETTING MARKET SYSTEMS Table 6.4: Record of beaten favourites when made favourite again on next start, by age group Age group
Winners
Total
Win % Profit/loss (£) Profit/loss (%)
Two-year-olds
407
1,006
40.5
26.69
Three-year-olds
387
1,078
35.9
-21.80
-2.0
22
56
39.3
-7.164
-12.8
919
3,098
29.7
-160.52
-5.2
1,735
5,238
33.1
-162.79
-3.11
Other age restricted Open to all ages Total
2.7
Eliminating false favourites
My research tells me that there are ‘true’ favourites and ‘false’ favourites. A ‘true’ favourite is one that has good, solid form. It should be difficult to identify any obvious holes in its form or to have any doubts about its well-being. On the other hand the ‘false’ favourite is the one that has a number of negatives against its name. I have found that the three biggest negatives are 1) wearing blinkers or a visor; 2) recently changed trainer or, 3) gelded since its last race.
The elimination of false favourites using these three simple factors seems to work. For instance, the record of beaten favourite two-year-olds can be improved significantly by eliminating those that were wearing blinkers or visors.
The presence of headgear is a powerful negative. This can be seen in Table 6.5. This shows that beaten favourite two-year-olds record a strike rate of 41 per cent if they have been made favourite again and are not wearing blinkers or visors. However, if they are wearing blinkers or visors then the strike rate falls alarmingly to just 26 per cent. I do not have data on those horses that
73
PROFITABLE BETTING SYSTEMS wear cheek pieces but I would reason that their record would be no more impressive than those horses wearing blinkers or visors. Table 6.5: Record of beaten favourite 2-year-olds considering whether or not they are blinkered or visored Type of headgear Winners Total % Win Never ran in blinkers/visor
391
945
41.4
First time in blinkers/visor
10
34
29.4
6
27
22.2
16
61
26.2
407
1,006
40.5
Blinkered/visored but ran before in b/v All blinkers/visors Total
Another negative factor is whether or not the horse has changed trainer since being a beaten favourite. This isn’t a very common occurrence when considering beaten two-year-old favourites that are starting favourite again but as you will see from Table 6.6 it is a definite negative. In my sample only 2 beaten favourites obliged from 12 runners that had recently changed trainer (16.7 per cent). Table 6.6: Record of beaten favourite 2-year-olds considering whether or not they have changed trainer Trainer Winners Total % Win Same trainer Change of trainer Total
405
994
40.7
2
12
16.7
407
1006
40.5
A further factor worth noting is whether or not the horse has recently been gelded. Recently gelded horses have an appalling win record, especially if they have been gelded as a two-year-old. This is hardly surprising when one considers the reasons as to why a colt might be gelded at such a tender age. Firstly, the colt is likely to be a bit of a rogue and has probably shown a tendency to have its mind on other things! Secondly, the decision has been made that the horse has no stallion potential. In other words connections
74
BETTING MARKET SYSTEMS believe that the horse is useless and that it wouldn’t be worth its keep as a sire. Thirdly, and perhaps a more plausible explanation, is that the physical effects of the operation, cause the horse to perform poorly.
Therefore the best way of making a modest profit from backing beaten favourites is to bet them in the following circumstances: 1.
Running in a race restricted to two-year-olds.
2.
Starting clear favourite, joint favourite or co favourite on their next start after being a beaten favourite.
3.
NOT wearing blinkers or visors.
4.
NOT changed trainer since last race.
5.
NOT recently gelded.
This simple system returns a modest profit and has a very high strike rate, making losing runs less likely. A profit of around five per cent can be earned from selections at a strike rate of almost 42 per cent. In terms of a comparison, if you backed every non-beaten favourite that also matched the above five criteria you would make a loss of around 7 per cent and return a strike rate of just 36 per cent (see Table 6.7). Table 6.7: Record of beaten favourite 2-year-olds (not changed trainer, gelded, or blinkered/visored) when starting favourite on next start Favourites Winners Total % Win Profit/loss (£) Profit/loss (%) Beaten favourite Non-beaten favourite
389
935
41.6
45.67
4.9
1,984
5,448
36.4
-360.20
-6.6
This system is not exactly one for raking in large amounts of money but it doesn’t take a great deal of time to identify a selection and it can be implemented using a daily racing paper. One is inevitably backing relatively short-priced picks (in my data sample the average price of each selection was
75
PROFITABLE BETTING SYSTEMS around 6/4). One also has to watch the betting market carefully to ensure that one actually bets the market favourite.
The moral of this tale? It sometimes pays to follow the crowd. However, it also begs a question: are there any other circumstances when it pays to follow the jolly? When to back the favourite In this section I want to look at the factors that help to identify winning and losing favourites on the flat.
There are a number of factors that differentiate between winning and losing favourites. One obvious factor is that Starting Price and odds-on favourites yield a lower level of loss than longer priced favourites. This reflects the longshot bias that I described earlier, whereby longer priced runners tend to represent poor value in relation to those at shorter odds.
A more original factor that divides winning and losing favourites is the gender of the horse. It is clear from the statistical evidence that male runners are a safer bet than female runners. Female favourites seem to be slightly more inconsistent, but this isn’t reflected in their odds. Female favourites return a greater level of loss than male favourites, with males making a loss of about 6.6 per cent on turnover and females losing nearly 9.5 per cent.
I’ve long held the opinion that it is wise to avoid favourites in top class races. The statistics back this up. My data shows that you would record a loss of over 11 per cent backing favourites in Group One and Group Two races. These races are always very competitive and the difference in ability between the runners is often, in my view, much smaller than it is in lower class races. It
76
BETTING MARKET SYSTEMS takes a truly exceptional horse to win a Group One race with any ease. However, at the other end of the class spectrum you often see one horse dominate a field. This particularly applies to maiden races. In these races you will see horses of widely different abilities, ranging from future Group performers to selling platers.
Maiden races provide fertile ground for favourite backers. It is in this form of race that the betting market is the most important guide to the performance of the runners. A bet on the market favourite would result in a loss of just 3.6 per cent while favourites in non-maiden races would result in a loss of nearly 9 per cent on turnover.
Handicaps are generally bad races for favourites. They are naturally competitive because the entry conditions stipulate that runners need to be within a certain ability range, as measured by official handicap ratings. Weight is also used to handicap the more able runners so, in theory at least, handicaps should equalise the chances of all the runners. This doesn’t happen, but you certainly see more close finishes in this kind of race than in any other, and so the favourite is really up against it. The bottom line is that you would lose more money backing favourites in handicaps so give them the swerve.
A horse’s headgear is another significant factor when it comes to differentiating between winning and losing favourites. A favourite that is wearing blinkers for the first time is definitely one to lay big time! The win rate of favourites wearing blinkers for the first time is only 28 per cent, which is much lower than for favourites in general. They make a huge loss of over 35 per cent.
77
PROFITABLE BETTING SYSTEMS I think Nick Mordin in Betting for a Living gave a very plausible explanation as to why horses wearing blinkers win infrequently. He said that blinkers work because they prevent horses from seeing behind. This makes them nervous of attack and their natural response is to flight. Consequently horses wearing blinkers, particularly when they have been applied for the first time, run as if someone has set them on fire. The poor animal though soon becomes exhausted from this outburst of nervous energy, and this probably explains why they have such a poor win record.
Table 6.8: Profit and loss of favourites, by horse and race characteristics Characteristic Profit/loss(%) Odds Odds-on
-6.25
Odds against
-7.55
Gender Male
-6.63
Female
-9.46
Class of race Group 1
-11.19
Group 2
-14.47
Lower class races
-7.48
Type of race Handicap
-9.94
Non handicap
-5.37
Maiden race
-3.60
Non maiden race
-8.73
78
BETTING MARKET SYSTEMS Characteristic
Profit/loss(%)
Headgear 1st time blinkers No blinkers
-35.23 -7.27
Clearly following favourites with any one of the more positive characteristics that I have described above is a better strategy than following favourites blindly.
However, the real money is to be found in laying the favourite on the exchanges if it has any of the negative characteristics in its profile. The most obvious strategy is to lay every favourite that is wearing blinkers for the first time. You are also on to a good thing if you lay the favourite in Group One and Group Two races. These are wide-open affairs and the market is a poor guide in such races. Female performers are also bad bets for favourite backers. They are often inconsistent and while they may have good recent form that is enough to make them favourite, there is no guarantee about how they are going to perform. I don’t want to be sexist but when it comes to horseracing sexism pays. Female favourites are best avoided.
The above all applies to flat racing, but what about the record of favourites in jump races? Are there any circumstances when it is better to back or to lay the favourite over the sticks?
Backing and laying favourites in hurdle races
When considering non-handicap hurdle races run in Great Britain and Ireland over the last 12 or so years what is striking is the fact that the record of the favourite varies enormously between different tracks.
79
PROFITABLE BETTING SYSTEMS
In Table 6.9 I have analysed the record of the favourite in non-handicap hurdle races by racecourse. As you can see there are no less than 25 tracks where the jolly favourite returns a profit.
Favourites at Cork do best in terms of rate of return, making a profit of nearly 17 per cent on turnover. The strike rate is also nearly one winner in every two selections. In contrast you wouldn’t want to be backing the favourite at courses such as Thurles, Kilbeggan, Sligo, Fairyhouse, Aintree, Ascot or Killarney where the favourite has a strike rate of less than 35 per cent and you would make a loss of more than 20 per cent on turnover. This definitely relates to my own experience of racing at Ascot. I’ve never done well at that track. I was all for them taking out the jump track and putting down an allweather surface!
The hurdles tracks at Cork, Listowel, Ballinrobe, Sandown, Wexford, Hereford, Haydock, Lingfield, Downpatrick, Huntingdon, Aintree, Windsor, Newbury, Taunton,
Clonmel,
Gowran
Park,
Worcester,
Cheltenham,
Navan,
Musselburgh, Perth, Stratford, Kelso, Southwell and Doncaster are the ones at which one should be backing the favourite.
Had you backed every favourite in non-handicap hurdles at these venues you would have made 4,096 bets and would have backed 1,862 winners (45 per cent). This would have resulted in a profit of around £219 or a rate of return of just over 5 per cent on all stakes invested. This isn’t a fortune by any means but it is a significant turnaround on the 5 per cent loss recorded by the favourite at all courses.
80
BETTING MARKET SYSTEMS Table 6.9: The record of the favourite in non-handicap hurdle races by racecourse, Great Britain and Ireland Course
Wins Runners %Win Profit/loss(£) Profit/loss(%)
CORK
53
107
49.5
16.62
15.53
LISTOWEL
22
51
43.1
6.40
12.55
BALLINROBE
15
31
48.4
2.88
9.29
SANDOWN
73
157
46.5
14.26
9.08
WEXFORD
29
72
40.3
6.09
8.46
HEREFORD
167
362
46.1
29.72
8.21
HAYDOCK
100
222
45.0
18.17
8.19
LINGFIELD
41
86
47.7
6.80
7.91
DOWNPATRICK
28
70
40.0
5.03
7.19
133
291
45.7
20.53
7.05
10
25
40.0
1.61
6.44
HUNTINGDON AINTREE WINDSOR
33
77
42.9
4.86
6.31
NEWBURY
102
240
42.5
15.13
6.30
TAUNTON
145
303
47.9
17.90
5.91
CLONMEL
36
89
40.4
4.47
5.02
GOWRAN PARK
33
77
42.9
3.31
4.30
WORCESTER
166
371
44.7
13.82
3.72
CHELTENHM
26
53
49.1
1.93
3.64
NAVAN
47
102
46.1
3.05
2.99
MUSSELBURGH
70
144
48.6
4.26
2.96
PERTH
121
247
49.0
7.29
2.95
STRATFORD
143
321
44.5
6.79
2.11
KELSO
119
256
46.5
3.98
1.55
SOUTHWELL
86
183
47.0
2.59
1.41
DONCASTER
64
156
41.0
0.51
0.33
KEMPTON
71
169
42.0
-1.60
-0.95
ROSCOMMON
15
36
41.7
-0.87
-2.42
CARTMEL
53
127
41.7
-3.51
-2.76
CATTERICK
71
173
41.0
-5.43
-3.14
LEOPARDSTOWN
55
128
43.0
-4.35
-3.40
BELLEWSTOWN
10
24
41.7
-0.96
-3.99
CARLISLE
75
167
44.9
-6.76
-4.05
81
PROFITABLE BETTING SYSTEMS Course
Wins Runners %Win Profit/loss(£) Profit/loss(%)
EXETER
83
187
44.4
-8.07
-4.32
LEICESTER
68
164
41.5
-7.15
-4.36
FOLKESTONE
75
173
43.4
-8.15
-4.71
UTTOXETER
183
427
42.9
-20.88
-4.89
HEXHAM
114
261
43.7
-12.77
-4.89
GALWAY
17
41
41.5
-2.11
-5.16
TRALEE
13
33
39.4
-1.82
-5.52
NEWCASTLE
83
195
42.6
-11.32
-5.80
SEDGEFIELD
128
317
40.4
-19.30
-6.09
CHEPSTOW
114
272
41.9
-17.18
-6.32
LIMERICK
58
144
40.3
-9.15
-6.35
PUNCHESTOWN
61
161
37.9
-10.61
-6.59
DOWN ROYAL
37
99
37.4
-7.37
-7.45
157
375
41.9
-29.72
-7.93
74
204
36.3
-18.45
-9.04
MARKET RASEN WARWICK LUDLOW CHELTENHAM
123
305
40.3
-27.71
-9.09
72
196
36.7
-18.07
-9.22
TIPPERARY
21
51
41.2
-5.00
-9.80
FAKENHAM
47
114
41.2
-11.42
-10.02
168
394
42.6
-41.93
-10.64
25
66
37.9
-7.41
-11.22
NEWTON ABBOT NAAS FONTWELL
156
402
38.8
-55.25
-13.74
TOWCESTER
101
277
36.5
-38.40
-13.86
WETHERBY
113
278
40.6
-38.56
-13.87 -13.95
AYR
79
178
44.4
-24.82
112
285
39.3
-43.02
-15.10
99
248
39.9
-39.43
-15.90
PLUMPTON
147
371
39.6
-61.37
-16.54
TRAMORE
23
77
29.9
-14.88
-19.33
THURLES
29
102
28.4
-20.80
-20.40
KILBEGGAN
23
68
33.8
-14.45
-21.25
SLIGO
16
50
32.0
-11.79
-23.57
FAIRYHOUSE
46
137
33.6
-33.27
-24.28
AINTREE
33
95
34.7
-23.17
-24.39
WINCANTON BANGOR
82
BETTING MARKET SYSTEMS Course ASCOT KILLARNEY
Wins Runners %Win Profit/loss(£) Profit/loss(%) 54
158
34.2
-42.54
-26.92
7
30
23.3
-17.08
-56.92
What about the record of favourites in chase races? Chase races In Table 6.10 below I have analysed the record of the favourite in nonhandicap chase races by course. As you can see there are no less than 13 tracks where the favourite returns a profit. Favourites at Thurles do best and return a profit of £21.84 to a £1 level stake, with more than 46 per cent of favourites returning to the winners’ enclosure. In contrast you wouldn’t want to be backing the favourite at courses such as Stratford, Kelso or Towcester. At these tracks the favourite has a strike rate of less than 40 per cent and above average losses would have been made.
The chase tracks at Thurles, Newcastle, Ayr, Sedgefield, Folkestone, Plumpton, Exeter, Wetherby, Huntingdon, Galway, Limerick, Cork, Leicester, Newbury, Windsor, Exeter, Bangor, Kilbeggan and Lingfield are the ones at which one should be backing the favourite. Table 6.10: The record of the favourite in non-handicap chase races by racecourse, Great Britain and Ireland Course Wins Runners %Win Profit/loss(£) Profit/loss(%) THURLES
30
65
46.2
21.84
33.60
NEWCASTLE
72
139
51.8
17.68
12.72
AYR
74
141
52.5
15.74
11.16
SEDGEFIELD
115
253
45.5
14.98
5.92
FOLKESTONE
83
171
48.5
11.52
6.74
PLUMPTON
82
162
50.6
11.11
6.86
EXETER
63
134
47.0
10.86
8.11
100
187
53.5
9.96
5.33
WETHERBY
83
PROFITABLE BETTING SYSTEMS Course HUNTINGDON
Wins Runners %Win Profit/loss(£) Profit/loss(%) 126
266
47.4
8.25
3.10
GALWAY
15
33
45.5
7.02
21.26
LIMERICK
26
57
45.6
6.17
10.82
CORK
25
55
45.5
5.59
10.17
LEICESTER
74
174
42.5
4.88
2.81
NEWBURY
60
129
46.5
4.59
3.56
WINDSOR
13
29
44.8
3.24
11.17
EXETER
50
110
45.5
2.04
18.56
BANGOR
92
184
50.0
1.83
0.99
KILBEGGAN
15
35
42.9
1.81
5.16
LINGFIELD
26
54
48.1
0.84
1.56
TIPPERARY
10
24
41.7
-0.21
-0.89
SANDOWN
76
182
41.8
-0.28
-0.15
KEMPTON
65
144
45.1
-0.66
-0.46
NAAS
14
35
40.0
-0.82
-2.34
105
227
46.3
-0.85
-0.37
PUNCHESTOWN
37
91
40.7
-0.87
-0.96
HAYDOCK
49
103
47.6
-0.97
-0.94
6
19
31.6
-1.35
-7.09
ASCOT
43
121
35.5
-1.47
-1.22
TRALEE
7
20
35.0
-1.67
-8.36
UTTOXETER
AINTREE
TRAMORE
16
38
42.1
-1.77
-4.66
5
15
33.3
-1.79
-11.94
21
55
38.2
-2.88
-5.24
SOUTHWELL
44
104
42.3
-3.48
-3.35
FAIRYHOUSE
31
85
36.5
-4.79
-5.63 -24.83
ROSCOMMON NAVAN
KILLARNEY
7
24
29.2
-5.96
19
50
38.0
-6.27
-12.54
5
22
22.7
-6.50
-29.55
CHELTENHM(NEW)
87
234
37.2
-7.19
-3.07
CLONMEL
18
58
31.0
-8.06
-13.89
CHEPSTOW
62
143
43.4
-8.15
-5.70
DOWNPATRICK
15
48
31.3
-8.59
-17.89
WEXFORD
10
37
27.0
-9.16
-24.74
CHELTENHAM(OLD) BALLINROBE
84
BETTING MARKET SYSTEMS Course
Wins Runners %Win Profit/loss(£) Profit/loss(%)
CATTERICK
39
104
37.5
-9.25
FAKENHAM
56
131
42.7
-10.47
-7.99
WORCESTER
80
192
41.7
-11.02
-5.74
LEOPARDSTOWN
28
74
37.8
-11.56
-15.63
FONTWELL
92
200
46.0
-11.58
-5.79
LISTOWEL
4
21
19.0
-11.97
-56.98
MARKET RASEN
108
237
45.6
-12.41
-5.24
GOWRAN PARK
12
45
26.7
-12.46
-27.68
MUSSELBURGH
30
78
38.5
-13.15
-16.85
PERTH
61
146
41.8
-13.20
-9.04
DONCASTER
36
92
39.1
-13.50
-14.67
HEXHAM
63
171
36.8
-13.72
-8.02
AINTREE
28
84
33.3
-14.41
-17.15
TAUNTON
48
123
39.0
-16.07
-13.06
WINCANTON
77
172
44.8
-16.13
-9.38
DOWN ROYAL
10
44
22.7
-17.97
-40.85
CARTMEL
21
64
32.8
-19.95
-31.18 -10.52
LUDLOW
-8.89
86
199
43.2
-20.93
110
228
48.2
-20.97
-9.20
HEREFORD
85
214
39.7
-21.56
-10.07
CARLISLE
47
119
39.5
-22.43
-18.85
WARWICK
76
175
43.4
-25.12
-14.35
STRATFORD
79
205
38.5
-25.24
-12.31
KELSO
79
201
39.3
-25.49
-12.68
TOWCESTER
78
201
38.8
-30.24
-15.04
NEWTON ABBOT
Why do favourites do better at some tracks than others?
It is anyone’s guess as to why the favourite is profitable to back at some tracks and not at others. However, it is probably worth speculating on a few theories. One hypothesis is that the on-course punters at these tracks are particularly skilled and only seriously bet horses that have a very good chance
85
PROFITABLE BETTING SYSTEMS of winning. This is certainly possible because during the study period oncourse punters largely determined the Starting Price returns. However, I find it hard to imagine that some racecourses attract more knowledgeable punters than others. There are, however, some people I know who do subscribe to this theory.
A more plausible theory, in my opinion, is that the favourite does better at some courses than others because of some characteristic about the track. For instance, novice races dominate the non-handicap chase category. In these races falls are generally common but it depends on the stiffness of the fences at a particular track. The favourite may therefore be less likely to fall at Thurles than at Warwick.
The fences, though, may be only one factor. It is a statistical fact that the favourite has a relatively poor record on heavy going and certainly some racecourses are more likely to produce soft or heavy going than others due to the local climate and drainage. Towcester, for example, always seems to race on bottomless ground, and the dip at the bottom of the uphill home straight is like a swamp after a downpour. This might explain why favourites have such a poor record at this track.
One could go on speculating about the reasons as to why favourites enjoy more success at some tracks than at others. However, writing out a betting slip isn’t the same as writing an essay. There are no prizes for well-argued answers. It is winners that count, or more accurately in the betting exchange era, it is winners and losers that count. You can make good money by backing the favourite in certain circumstances and opposing them in others.
86
Chapter 7: Pedigree Profits I have always found races for two-year-olds fascinating. I know this is not a view shared by most serious backers, who prefer betting in races where all the runners have exposed form, but there is something exciting about the prospect of watching a race featuring a number of previously unraced juveniles. One of them might be a future Classic or Group race winner, or one might even develop into a racing great like Mill Reef or Dancing Brave.
As a spectacle, juvenile races certainly have an appeal and I have been cheered up on many a visit to low grade meetings, at places like Yarmouth or Leicester, by watching one of the big stables race one of their more promising two-year-olds for the first time.
However, as a betting medium, races for juveniles are not for the unwary. The statistics reveal that if you backed a horse at random in every juvenile contest you would lose over 30 pence for every pound staked. This is considerably more than for other types of race and reflects the fact that juvenile races are particularly difficult to work out. Profiting from pedigree data
Despite the inherent difficulties of backing two-year-olds I have found that a profit can be made by studying horse’s pedigree, and systems can be developed around this theme. After all, a horse’s genetic inheritance is probably the most important of all factors. A horse’s genes determine its level of class, its stamina, temperament and early speed.
PROFITABLE BETTING SYSTEMS
Most punters ignore a horse’s bloodline when studying the form either because they do not believe the information to be important or because they cannot obtain basic information on a horse's sire, dam and grandsire. On the latter point I have noted that most of the national newspapers do not even provide basic pedigree information for juvenile races. Fortunately the racing press provides detailed data but the problem for punters is to understand its meaning. For instance, if a juvenile is running for the first time in a five furlong contest and is sired by Grand Lodge is this a positive or a negative?
The Racing Post does offer some qualitative information as to whether a particular sire produces juvenile winners or not, and provides some information on whether a runner is likely to be suited to a particular distance or going. However, the punter wanting to compare the merit of one runner against another based on its pedigree, needs hard statistical data on each horse’s sire at the very least. For instance, with betting purposes in mind, it will be important to know the strike rate of a sire with its juvenile runners over different distances so that one can rate the relative chance of each runner. This sort of information is not readily available and requires more in-depth research.
Researching sire records
I have studied the records of all sires that had juvenile runners in Great Britain and Ireland in recent years, by race distance.
In the Tables that follow I have identified those sires that recorded a level stakes profit with their offspring during the study period. In order to qualify for
88
PEDIGREE PROFITS inclusion in the Tables a sire had to have recorded at least 30 runners. I set this cut-off in order to provide more reliable findings.
Table 7.1 shows the record of sires that had juvenile runners in five-furlong races. The sire to follow in this category is Pivotal. In the study period he had 95 runners and 25 winners. This represents a strike rate of over 26 per cent. His offspring are also unfashionable among fellow punters and many are allowed to go off at generous odds. Consequently backing all the offspring of Pivotal in five-furlong races would have netted you a profit of £95 to a £1 stake.
Over a distance of five furlongs the progeny of Grand Lodge and Intikhab were also worth following.
At the bottom end of the Table the sires Spectrum and Dolphin Street both return a profit but their strike rate is disappointing and I would be cautious about backing their offspring over the minimum trip.
Table 7.1: Record of sires that had juvenile runners in five-furlong races Sire PIVOTAL
Wins Runners % Win Profit/loss (£) Profit/loss (%) 25
95
26.3
95.12
8
37
21.6
24.25
65.54
INTIKHAB(USA)
11
55
20.0
15.98
29.06
CAPE CROSS(IRE)
GRAND LODGE(USA)
100.13
16
85
18.8
18.69
21.99
DANSILI
8
47
17.0
30.00
63.83
TRANS ISLAND
9
54
16.7
28.92
53.55
NICOLOTTE
6
38
15.8
14.50
38.16
GREAT COMMOTION(USA) BAHAMIAN BOUNTY
5
35
14.3
15.16
43.31
21
152
13.8
45.61
30.01
DAGGERS DRAWN(USA)
14
107
13.1
6.20
5.80
89
PROFITABLE BETTING SYSTEMS Sire
Wins Runners % Win Profit/loss (£) Profit/loss (%)
GREENSMITH
7
54
13.0
16.03
29.69
TITUS LIVIUS(FR)
22
170
12.9
58.88
34.63
ORPEN(USA)
15
116
12.9
43.75
37.72
4
33
12.1
16.25
49.24 14.02
LAHIB(USA)
4
33
12.1
4.63
SRI PEKAN(USA)
TOMBA
11
93
11.8
0.25
0.27
MAKBUL
14
121
11.6
15.08
12.46
PARIS HOUSE
17
150
11.3
4.13
2.76
DARNAY
4
37
10.8
4.92
13.29
MOST WELCOME
6
59
10.2
28.75
48.73
SPECTRUM(IRE)
5
51
9.8
29.21
57.27
DOLPHIN STREET(FR)
3
38
7.9
12.00
31.58
At the six furlongs distance there are a number of sires that record fantastic strike rates with their offspring.
Seeking the Gold, Storm Cat and Danzig do superbly well, year after year, and regularly record a strike rate in and around 30 per cent.
In the study period Danzig and Danehill also did very well with their offspring; sadly though both died a couple of years ago. There may be other sires listed in the Tables that are also recently deceased or have recently been retired from stud duties. This is a problem with compiling lists of sires to follow and one needs to keep records up-to-date. It is well worth keeping an eye on the bloodstock pages in the racing press to note any sires that have gone out of business.
Table 7.2: Record of sires that had juvenile runners in six-furlong races Sire SEEKING THE GOLD(USA)
Wins Runners % Win Profit/loss (£) Profit/loss (%) 14
41
90
34.1
3.86
9.41
PEDIGREE PROFITS Sire
Wins Runners % Win Profit/loss (£) Profit/loss (%)
STORM CAT(USA)
14
46
30.4
12.68
27.57
DANZIG(USA)
17
57
29.8
13.11
23.01
DISTANT RELATIVE
11
45
24.4
34.38
76.39
RAHY(USA)
13
57
22.8
21.40
37.55
DANEHILL(USA)
46
205
22.4
10.61
5.17
ANABAA(USA)
8
38
21.1
20.75
54.61
MEDICEAN
6
30
20.0
31.50
105.00
NICOLOTTE
8
42
19.0
22.00
52.38
CADEAUX GENEREUX
29
166
17.5
26.27
15.82
DANZERO(AUS)
10
63
15.9
45.70
72.54 52.43
PIVOTAL
31
213
14.6
111.68
SRI PEKAN(USA)
21
145
14.5
12.57
8.67
5
35
14.3
4.20
12.00
CHARNWOOD FOREST(IRE) RED RANSOM(USA)
18
130
13.8
5.32
4.09
8
58
13.8
13.76
23.72
TRANS ISLAND
10
74
13.5
22.92
30.97
SPECTRUM(IRE)
19
143
13.3
58.07
40.61
NIGHT SHIFT(USA)
39
299
13.0
38.42
12.85
DR DEVIOUS(IRE)
7
54
13.0
18.94
35.07
CELTIC SWING
5
43
11.6
3.25
7.56
REPRIMAND
4
35
11.4
33.00
94.29
DISTANT MUSIC(USA)
MUJAHID(USA)
11
98
11.2
1.06
1.08
TITUS LIVIUS(FR)
15
159
9.4
30.75
19.34
ZILZAL(USA)
7
78
9.0
40.63
52.08
ROYAL ACADEMY(USA)
5
56
8.9
33.00
58.93
VICTORY NOTE(USA)
6
71
8.5
10.01
14.10
AMONG MEN(USA)
2
33
6.1
69.00
209.09
The seven-furlong plus race category is interesting because distances of seven furlongs or more are really more about stamina whereas five and six furlong contests, that dominate the racing calendar in the early part of the season, are all about raw speed. Consequently the key sires for seven-furlong contests (see Table 7.3) are different from the ones listed in Tables 7.1 and
91
PROFITABLE BETTING SYSTEMS 7.2. Two-year-olds that can win over seven furlongs often go onto become milers and middle distance performers at three.
The sires Common Grounds and Fusaichi Pegasus have an excellent winnerto-runners ratio at seven furlongs, and decent profits would have been made backing their offspring. The same could also be said of offspring from King’s Best, Tagula, Bachir and Intikhab.
Again one has to be careful about some of the sires listed in Table 7.3. A number recorded terrible strike rates (i.e. below 10 per cent) but record profits because they returned a couple of winners at long odds. These sires should be avoided.
Table 7.3: Record of sires that had juvenile runners in seven-furlong races Sire
Winners Runners % Win Profit/loss (£) Profit/loss (%)
COMMON GROUNDS
7
30
23.3
16.50
55.00
FUSAICHI PEGASUS(USA) KING`S BEST(USA)
8
35
22.9
29.60
84.57
27
140
19.3
7.96
5.69
TAGULA(IRE)
27
145
18.6
38.79
26.75
BACHIR(IRE)
6
33
18.2
4.00
12.12
INTIKHAB(USA)
19
108
17.6
40.38
37.38
BARATHEA(IRE)
52
330
15.8
9.85
2.98
EL PRADO(IRE)
5
32
15.6
16.00
50.00 154.34
BOLD EDGE
5
33
15.2
50.93
MONTJEU(IRE)
21
141
14.9
9.46
6.71
GALILEO(IRE)
10
71
14.1
18.32
25.80
BISHOP OF CASHEL
12
88
13.6
78.38
89.06
4
30
13.3
10.75
35.83
EXIT TO NOWHERE(USA) FRUITS OF LOVE(USA)
11
84
13.1
0.38
0.45
HALLING(USA)
23
177
13.0
9.30
5.26
ELNADIM(USA)
4
32
12.5
9.75
30.47
92
PEDIGREE PROFITS Sire COZZENE(USA)
Winners Runners % Win Profit/loss (£) Profit/loss (%) 6
48
12.5
5.47
11.41
MISTER BAILEYS
16
133
12.0
15.15
11.39
DESERT STYLE(IRE)
17
160
10.6
6.78
4.24
7
66
10.6
7.33
11.11 0.30
MOZART(IRE) SECOND EMPIRE(IRE)
10
96
10.4
0.28
BAHHARE(USA)
18
176
10.2
10.42
5.92
8
81
9.9
9.10
11.23
26
268
9.7
72.45
27.03
6
67
9.0
8.50
12.69
DR FONG(USA)
16
179
8.9
6.02
3.36
ROSSINI(USA)
8
90
8.9
83.00
92.22
FANTASTIC LIGHT(USA)
7
79
8.9
28.20
35.70
CROCO ROUGE(IRE)
5
57
8.8
19.63
34.43
KALANISI(IRE)
6
70
8.6
20.00
28.57
MILLKOM
4
48
8.3
17.75
36.98
JOSR ALGARHOUD(IRE)
9
112
8.0
24.69
22.05
GENEROUS(IRE)
3
38
7.9
61.00
160.53
INZAR(USA)
7
90
7.8
37.50
41.67
BERTOLINI(USA)
6
78
7.7
51.57
66.12
DANZERO(AUS)
4
59
6.8
31.00
52.54
DOCKSIDER(USA) VETTORI(IRE) MAKBUL
In terms of a system it is clear from the above Tables that a number of sires have an impressive win record with their two-year-old offspring at certain distances. In particular it is well worth following runners sired by Pivotal, Grand Lodge and Intikhab at five furlongs.
In six furlong races Seeking the Gold, Storm Cat, Distant Relative, Rahy, Anabaa and Medicean are the ones to follow.
In contests that place a greater demand on stamina (i.e. seven furlongs and more) Common Grounds and Fusaichi Pegasus produce plenty of winners between them and return a decent profit.
93
PROFITABLE BETTING SYSTEMS In following this group of elite sires you may occasionally find two or more selections in the same race. You can deal with this in a number of ways. For instance, you could bet on the horse whose sire has the highest winners-torunners ratio recorded in the above Tables. This is a reasonable strategy but in the last few flat race seasons I have found it more profitable to back every selection each-way.
Backing the juvenile runners of certain sires is just one way to make a profit out of pedigree analysis. There are a number of other pedigree-based systems that are well worth adding to your punting arsenal.
Don’t forget the horse’s birthday: the significance of foaling dates
When betting in juvenile races, one of the most significant angles that I have found is the horse’s birthday, or more specifically the month in which it was foaled. In Table 7.4 I have re-produced my analysis of the win record of twoyear-olds by the month in which they were foaled.
Table 7.4: Win record of juveniles that raced in Great Britain and Ireland, by month foaled Month foaled
Winners Runners % Win
JAN
164
1,366
12.0
FEB
425
4,031
10.5
MAR
479
5,458
8.8
APR
455
5,587
8.1
MAY
103
1,760
5.9
JUN
1
42
2.4
DEC
-
3
0.0
Total
1,627
18,247
8.9
94
PEDIGREE PROFITS The data presented in Table 7.4 clearly demonstrates a significant bias in favour of two-year-olds that were born in the early part of the year. For instance, juveniles that were born in the previous January won at a rate of 12 per cent but horses foaled in June won at a rate of 2.4 per cent. In other words juveniles foaled in January were five times more likely to win than those foaled in June.
This result is not surprising when one considers the administration of horse racing in Great Britain and Ireland, and how the authorities calculate a horse’s racing age.
In these countries all racehorses share the same birthday, namely 1st January. This makes it easier for the authorities to calculate a horse’s age because they do not have to find out the date of the horse’s actual birthday. It does though create a major bias in two-year old races because it means that a horse born on 1st June 2006 would be classified as a juvenile eligible for racing in January 2008, when their actual chronological age would be 19 months.
In comparison a horse actually born on 1st January 2006 would be 24 months on 1st January 2008, giving it an age advantage of five months over its more immature counterpart. Most trainers do not bother running June foals until they are officially three years of age in order to overcome this bias, and in the current sample few bothered to race horses that were born from June onwards.
The data in Table 7.4 clearly suggests that when betting in juvenile races one needs to concentrate on those runners that were foaled between the months
95
PROFITABLE BETTING SYSTEMS of January and April, with preference being for those foaled in the months of January and February.
You will not find foaling date information in the racing pages of the main newspapers. You will have to obtain this information from the racing press or some other specialist source.
However, foaling date information on its own would not be sufficient to produce a profitable betting system, but this outlook is transformed when one bets on those sires that have a decent record with their juvenile runners, and on those runners which were born before the month of May.
I have studied the records of all sires that had juvenile runners in Great Britain and Ireland over the last five years, and had runners with birthdays in the months of January, February, March and April.
In Tables 7.5 I have identified those sires that recorded a level stakes profit with their offspring during the study period.
Table 7.5: Record of sires that had juvenile runners, born between January and April Sire
Wins Runners % Win Profit/loss (£) Profit/loss (%)
AIR EXPRESS(IRE)
14
45
31.1
48.89
108.63
MUJTAHID(USA)
11
40
27.5
40.07
100.18
FUSAICHI PEGASUS(USA)
12
46
26.1
32.02
69.60
DISTANT RELATIVE
23
106
21.7
36.63
34.55
9
50
18.0
17.83
35.67
CADEAUX GENEREUX(GER) INTIKHAB(USA)
40
240
16.7
20.31
8.46
KYLLACHY
6
36
16.7
1.08
3.01
ALJABR(USA)
6
36
16.7
0.44
1.23
96
PEDIGREE PROFITS Sire LOUP SAUVAGE(USA) PIVOTAL COZZENE(USA)
Wins Runners % Win Profit/loss (£) Profit/loss (%) 8
49
16.3
1.65
3.37
75
481
15.6
109.80
22.83
9
62
14.5
23.33
37.62
NICOLOTTE
16
112
14.3
20.00
17.86
HALLING(USA)
33
235
14.0
7.30
3.11
7
50
14.0
9.75
19.50
25
179
14.0
14.60
8.16
6
43
14.0
10.00
23.26
EXIT TO NOWHERE(USA) RAHY(USA) EL PRADO(IRE) SUPERIOR PREMIUM
9
66
13.6
20.25
30.68
RED RANSOM(USA)
23
174
13.2
5.61
3.22
GALILEO(IRE)
10
78
12.8
11.32
14.51
FANTASTIC LIGHT(USA)
14
111
12.6
24.61
22.17
8
64
12.5
8.41
13.14
MEDICEAN
13
106
12.3
4.75
4.48
BISHOP OF CASHEL
23
192
12.0
43.23
22.51
DANZERO(AUS)
20
169
11.8
62.86
37.19
TRANS ISLAND
23
198
11.6
9.33
4.71
GREENSMITH
11
96
11.5
12.78
13.32
8
71
11.3
9.14
12.88
TITUS LIVIUS(FR)
52
480
10.8
65.53
13.65
BERTOLINI(USA)
33
307
10.7
5.92
1.93
OBSERVATORY(USA)
18
176
10.2
2.83
1.61
KINGSINGER(IRE)
4
44
9.1
14.07
31.98
AUCTION HOUSE(USA)
3
34
8.8
21.57
63.45
KALANISI(IRE)
7
81
8.6
16.50
20.37
CROCO ROUGE(IRE)
5
62
8.1
14.63
23.59
GENEROUS(IRE)
4
57
7.0
43.73
76.71
16
236
6.8
3.32
1.41
AMONG MEN(USA)
5
81
6.2
43.50
53.70
OVERBURY(IRE)
2
34
5.9
5.50
16.18
GREAT COMMOTION(USA)
MILLKOM
ROSSINI(USA)
97
PROFITABLE BETTING SYSTEMS In my opinion the final results are staggering. Air Express, Mujtahid, Fusaichi Pegasus, and Distant Relative all record a strike rate of over 20 per cent and made a pretty decent return between them. These results are impressive when they are put into context because two-year-olds born between January and April win at a rate of only 10 per cent, and make a loss of about 30 per cent on turnover.
Air Express does particularly well with his runners. He had 45 runners in the sample period and recorded 14 winners. He did not appear to be a fashionable sire with other punters because his runners tended to go off at decent odds. As a result you would have made a profit of nearly £49 to a £1 stake betting on his runners. This represents a rate of return of nearly 109 per cent.
Mujtahid (by Woodman) is another sire worth following, especially with his runners over five and six furlongs. This won’t come as a surprise to anyone who remembers Mujtahid in his racing days. He was probably one of the fastest two-year-old sprinters that I ever saw. He averages a strike rate of nearly 28 per cent and you would have made over £40 betting on his runners. This translates into a rate of return of over 100 per cent.
Next in the list is Fusaichi Pegasus and you would need to back plenty of winners to be able to afford one of his sons at auction because his offspring have been selling for millions recently, with Coolmoore splashing out a record $10m for one of his yearlings a couple of years ago. He is certainly a sire well worth following because his runners have a strike rate of over 26 per cent. The betting public also hadn't caught onto his ability because the average Starting Price of his runners was over 5 to 1 during the study period.
98
PEDIGREE PROFITS Distant Relative does not do as well as some of the other sires already mentioned but he is also well worth following.
There are plenty of sires listed in Table 7.5 that make decent profits but I would not recommend following any sire that had a strike rate of less than 15 per cent because losing runs would be too long and too expensive. This rule still gives 10 sires to follow during the juvenile racing season. This elite group of sires had 1,129 runners and 204 winners during the study period, giving them a strike rate of 18 per cent and a profit of £309 to a £1 stake (a return of 27 per cent).
Clearly the above findings show that some sires in flat racing are well worth following, but what about National Hunt sires? National Hunt sires to follow
In jump racing the number of bettors who bother to study a horse’s pedigree are even fewer in number than in flat racing.
The conventional wisdom is that ‘jumpers’ have raced so many times and in so many different circumstances that their distance and going preferences are established and well known. This is certainly true in most cases and there is probably not a lot to learn about a handicap chaser that has run more than one hundred times. A study of that horse’s breeding is unlikely to tell you anything that you didn’t already know about the horse by simply reading its previous form.
However, there will be times when a chaser or hurdler will be running in conditions that it had not previously experienced, or had encountered on only
99
PROFITABLE BETTING SYSTEMS a limited number of previous occasions. A reasonable knowledge of a horse’s pedigree can be very useful in this kind of situation, and can pay rich dividends.
In the next section I want to explain to you how you can make a decent profit by following the runners of certain National Hunt sires when their offspring are running on heavy ground. The effect of the going in jump races
The underfoot conditions play a major part in determining the outcome of a horserace, and in extreme conditions the horse that wins is often the one that was better able to handle those conditions.
When the official going is described as ‘heavy’ the formbook can often go out of the window and long shots often prevail. The winner will be the one that has the stamina and the endurance to squelch through the ground. This information is not necessarily to be found in the form book because genuinely heavy ground (not simply soft or yielding) is relatively rare. The clue as to how a horse will act on this type of going is to be found in the horse’s bloodline.
In Table 7.6 I have identified those jump sires that recorded a level stakes profit with their offspring when racing on heavy ground.
The French bred sire Cadoudal heads the Table. His offspring have a terrific record when running on heavy ground. In the study period he had 102 runners and recorded 29 winners. This represents a strike rate of more than 28 per cent.
100
PEDIGREE PROFITS Followers of Cadoudal would have been happy with their £54 profit to a level £1 stake. This would have resulted in a rate of return of just over 53 per cent. His runners tend to be returned at short-odds but in recent seasons he had three nice priced winners in Cadogan at 7 to 2, Ivan de Vonnas also a 7 to 2 winner and Bumper at 4 to 1.
The sire Useful though did have plenty of runners during the last season, and his offspring really excel when the mud is flying. In the study period he clocked up 50 runners on heavy ground and had 12 winners. This gave him nearly one winner for every four runners. The more impressive statistic though is his rate of return of 98 per cent. In other words you would have doubled your money backing his runners. The rate of return was boosted significantly by Kercabellecs’ victory at Market Rasen on 26th March 2006. The ground that day was bottomless but as one would have expected from a son of Useful he coped well with the conditions and obliged at odds of 25 to 1.
The German based sire Acatenango is well known for producing offspring that are suited by a bit of cut in the ground and I recall that Acatenango was himself well suited to soft ground conditions. His record of 7 winners from 33 runners in jump races run in heavy going is therefore hardly surprising, but because his record with runners on heavy ground is so well known the odds of reward are lower than those recorded by some of the other sires featured in the Table. He is though, capable of producing a winner at double figure odds.
Another sire of interest is Buckskin. He is the sire of Ebony Light and when the ground came up bottomless at Haydock on 21st January 2006 those who remembered that Buckskin recorded a strike rate of over 19 per cent with his runners on heavy ground were rewarded. Ebony Light bolted up at odds of 33 to 1!
101
PROFITABLE BETTING SYSTEMS Further down the list is Karinga Bay. He is a real stamina influence and he does particularly well with his runners when they are running on heavy ground, as evidenced by his record of 23 winners from 132 runners (a strike rate of 17.4 per cent). His runners though tend to be ignored by other punters and that allowed his followers to earn a level stake profit of nearly £58 to a £1 stake during the survey.
Table 7.6: Record of National Hunt sires that had runners in Great Britain and Ireland that raced on heavy ground Sire
Wins Runs % Win Profit/loss (£) Profit/loss (%)
CADOUDAL(FR)
29
102
28.4
54.31
53.25
USEFUL(FR)
12
50
24.0
49.11
98.22
ACATENANGO(GER)
7
33
21.2
1.97
5.97
NIKOS
7
35
20.0
0.78
2.24
51
263
19.4
75.95
28.88
BUCKSKIN(FR) THATCHING ELA-MANA-MOU
8
42
19.0
23.88
56.85
15
79
19.0
14.70
18.61 11.05
CASTLE KEEP
11
59
18.6
6.52
ROLFE(USA)
10
54
18.5
4.75
8.80
BROKEN HEARTED
12
66
18.2
34.65
52.50
TOP VILLE
8
44
18.2
1.53
3.47
HIGHEST HONOR(FR)
7
39
17.9
6.83
17.52
HIGH ESTATE
9
51
17.6
32.05
62.84
10
57
17.5
6.00
10.53
DESERT KING(IRE)
7
40
17.5
17.00
42.50
ALIAS SMITH(USA)
7
40
17.5
14.50
36.25
SANGLAMORE(USA)
7
40
17.5
10.68
26.69
CALLERNISH
15
86
17.4
6.81
7.92
KARINGA BAY
23
132
17.4
57.82
43.80
KING LUTHIER
10
58
17.2
26.75
46.12
BOB’S RETURN(IRE)
NASHWAN(USA)
16
94
17.0
137.38
146.14
ZILZAL(USA)
7
43
16.3
31.75
73.84
KADALKO(FR)
6
37
16.2
1.88
5.07
102
PEDIGREE PROFITS Sire
Wins Runs % Win Profit/loss (£) Profit/loss (%)
AKARAD(FR)
5
31
16.1
22.00
SHERNAZAR
15
95
15.8
4.37
4.60
5
32
15.6
3.75
11.72
10
64
15.6
1.75
2.73
GLENSTAL(USA) BUSTINO DEFINITE ARTICLE
70.97
6
40
15.0
16.26
40.65
MIDYAN(USA)
10
68
14.7
12.88
18.93
LEAR FAN(USA)
11
76
14.5
0.82
1.08
LYPHENTO(USA)
5
35
14.3
8.75
25.00
12
85
14.1
11.20
13.17
8
57
14.0
34.50
60.53
SIMPLY GREAT(FR) DR MASSINI(IRE) SAFETY CATCH(USA)
6
44
13.6
20.25
46.02
RAINBOWS FOR LIFE(CAN)
6
46
13.0
7.09
15.42
TEAMSTER
7
55
12.7
3.73
6.78
LAW SOCIETY(USA)
8
64
12.5
5.38
8.40
WITNESS BOX(USA)
9
73
12.3
86.25
118.15
EL CONQUISTADOR
11
90
12.2
5.88
6.53
WOODMAN(USA)
5
41
12.2
13.83
33.74
PURSUIT OF LOVE
6
50
12.0
0.80
1.60
SIR HARRY LEWIS(USA)
8
69
11.6
108.53
157.29
TREASURE HUNTER
8
69
11.6
19.00
27.54
DESTROYER
5
46
10.9
6.50
14.13
12
114
10.5
22.96
20.14
AFZAL
RUDIMENTARY(USA)
7
67
10.4
23.82
35.55
LIR
6
59
10.2
42.00
71.19
RA NOVA
9
89
10.1
20.50
23.03
BE MY CHIEF(USA)
7
70
10.0
49.50
70.71
FAUSTUS(USA)
7
70
10.0
8.95
12.79
ROYAL VULCAN
6
63
9.5
2.50
3.97
26
280
9.3
21.21
7.57
MAGICAL WONDER(USA)
6
67
9.0
4.75
7.09
GREEN DANCER(USA)
3
34
8.8
35.25
103.68
BALLACASHTAL(CAN)
5
57
8.8
22.50
39.47
PRESIDIUM
10
117
8.5
7.75
6.62
REPRIMAND
4
50
8.0
6.00
12.00
COMMANCHE RUN
103
PROFITABLE BETTING SYSTEMS Sire
Wins Runs % Win Profit/loss (£) Profit/loss (%)
DON’T FORGET ME
6
77
7.8
0.75
0.97
SHARP DEAL
1
32
3.1
69.00
215.63
At the bottom of Table 7.6 several sires return a profit when their offspring are running on heavy ground but record a terrible strike rate. Sharp Deal and Don’t Forget Me, record a profit but have a strike rate of less than 8 per cent. I would recommend following any sire listed above Midyan, ignoring the rest. Hurdle and chase debutantes
Another profitable angle is to study the record of sires when they have a runner that is making its debut in either a chase or a hurdle race. In Table 7.7 I have identified those sires that recorded a level stakes profit with their offspring when they had their first run in a chase, and in Table 7.8 I have done the same but this time for runners running for the first time in a hurdle race.
Chasers
Tragic Role recorded the highest winning strike rate of all the sires listed in Table 7.7. His 34 chase debutantes recorded 9 victories. In other words more than one-quarter of all his runners returned to the winners enclosures after their first run in a chase. The odds on the winning selections were a bit on the skinny side and you would have earned a profit of just over £3 had you backed all his runners, but the rate of return was still around 10 per cent.
Petardia recorded proportionally fewer winners than Tragic Role (seven winners from 37 runners) but the level stakes profit was £9.75 and the rate of return was an impressive 26 per cent.
104
PEDIGREE PROFITS Next in the list comes Un Desperado. He is probably well known to most punters as he is the sire of the ill-fated triple Cheltenham Gold Cup winner Best Mate amongst other top-class chasers. His record of 21 winners from 127 runners is impressive (16.5 per cent) and despite his popularity as a sire to follow his runners returned a reasonable level of profit.
Table 7.7: National Hunt sires record with runners competing in a chase for the first time Sire
Wins Rnrs % Win Profit/loss (£) Profit/loss (%)
TRAGIC ROLE(USA)
9
34
26.5
3.36
9.88
PETARDIA
7
37
18.9
9.75
26.35
UN DESPERADO(FR)
21
127
16.5
14.18
11.17
GILDORAN
11
68
16.2
12.70
18.68
KEEN
5
31
16.1
0.50
1.61
KRIS
11
70
15.7
13.22
18.88
SIR HARRY LEWIS(USA)
5
33
15.2
38.23
115.84
BE MY GUEST(USA)
5
34
14.7
48.25
141.91
ROSELIER(FR)
42
287
14.6
10.64
3.71
GUNNER B
12
91
13.2
138.28
151.95
SHERNAZAR
5
38
13.2
17.07
44.92
SULA BULA
6
46
13.0
17.00
36.96
BEAU SHER(IRE)
3
30
10.0
5.00
16.67
SURE BLADE(USA)
3
30
10.0
1.00
3.33
PRESIDIUM
5
61
8.2
22.50
36.89 4.55
YASHGAN
5
77
6.5
3.50
PRIMITIVE RISING(USA)
6
101
5.9
3.98
3.94
HOMO SAPIEN
2
36
5.6
20.00
55.56
Roselier is another sire recognisable to many astute punters because he is a regular source of top class chasers and therefore it is hardly surprising that he recorded 42 winners from 287 runners (14.6 per cent). The odds of reward on his runners though are, like for Un Desperado, pretty short because he is well
105
PROFITABLE BETTING SYSTEMS exposed as a sire of decent chasers. Furthermore, his offspring often find themselves in the care of the most high profile trainers and this does limit the opportunity for striking a value bet, but they are always worthy of respect.
Punters don’t seem to have worked out that Sir Harry Lewis and Be My Guest are top-class sires of chasers and they both record a higher proportion of winners than Roselier. Consequently they are far more profitable to follow. For instance, Sir Harry Lewis recorded a rate of return of well over 100 per cent and so did Be My Guest.
At the bottom end of Table 7.7 there are a few sires whose runners I would avoid. I certainly wouldn’t bother to follow the offspring of Homo Sapien, Primitive Rising or Yashgan. They all have a strike rate of below seven per cent and while they make a profit this is down to the fact that they have each recorded the odd long priced winner. Generally speaking I probably wouldn’t bother to follow any sire that didn’t record a win strike rate of around 15 per cent or better.
Hurdlers
Table 7.8 details the record of National Hunt sires that had a runner competing in a hurdle race for the first time in the last few years.
The names listed are very different from those listed in Table 7.7, with only Be My Guest and Beau Sher featuring in both Tables. This reflects the fact that chasers and hurdlers are very different beasts.
106
PEDIGREE PROFITS It takes a certain kind of horse to win a chase and a completely different sort of animal to be successful over hurdles. Hurdle racing is about speed whereas chasing is more about stamina. Flat-bred horses therefore tend to do well over hurdles. This is illustrated in Table 7.8 by the number of flat race sires that have a good record with hurdlers. For instance, Mark of Esteem, Unfuwain, Darshaan, Machiavellian, Groom Dancer and Woodman are all sires of topclass horses on the flat, and all were successful in that sphere when they themselves were racing. You would do very well though to follow these same sires when they have a runner that is running in a hurdle race for the first time.
Royal Charter and Beau Sher had the highest ratio of winners to runners in the survey. Royal Charter had eight winners from 35 runners during the study period (22.9 per cent) and Beau Sher had an almost identical record. Royal Charter though was far more profitable to follow because his runners netted a return of over £50 to a £1 stake whereas the offspring of Beau Sher made a modest £8.19.
Table 7.8: National Hunt sires record with runners competing in a hurdle race for the first time Sire
Wins Rnrs % Win Profit/loss (£) Profit/loss (%)
ROYAL CHARTER(FR)
8
35
22.9
50.50
144.29
BEAU SHER(IRE)
8
37
21.6
8.19
22.15 68.10
MARK OF ESTEEM(IRE) UNFUWAIN(USA) BE MY GUEST(USA)
6
35
17.1
23.83
10
59
16.9
56.36
95.53
6
36
16.7
59.90
166.39
DARSHAAN
9
61
14.8
13.27
21.75
WITNESS BOX(USA)
6
43
14.0
80.00
186.05
GROOM DANCER(USA)
6
45
13.3
1.00
2.22
RELIGIOUSLY(USA)
7
53
13.2
3.55
6.70
POLAR FALCON(USA)
6
47
12.8
5.00
10.64
MACHIAVELLIAN(USA)
5
42
11.9
15.50
36.90
107
PROFITABLE BETTING SYSTEMS Sire
Wins Rnrs % Win Profit/loss (£) Profit/loss (%)
DEFINITE ARTICLE
7
65
10.8
12.25
18.85
WOODMAN(USA)
3
30
10.0
18.00
60.00
ACATENANGO(GER)
3
33
9.1
28.88
87.50
TEENOSO(USA)
8
95
8.4
5.25
5.53
DENEL(FR)
3
44
6.8
12.00
27.27
EMARATI(USA)
1
30
3.3
21.00
70.00
PICCOLO
1
36
2.8
65.00
180.56
I definitely wouldn’t follow runners sired by Piccolo or Emarati because while they each recorded an impressive level of profit this was attributable to a couple of long priced winners.
Those sires with a strike rate of 10 per cent and upwards are the ones to follow in my view. You may wonder why I suggested a cut-off figure of around 15 per cent for chase sires to follow but only 10 per cent for hurdlers? I’m not trying to be inconsistent but simply adjusting for the fact that field sizes in hurdle races are generally much higher than for chasers.
108
Chapter 8: Trainers to follow The more I have studied and betted on horse racing I have come to realise that the competence of a horse’s trainer is vital. I’m not going to win any prizes for originality with such an obvious statement but it is sometimes worthwhile stating the obvious because you can make decent profits by making a detailed analysis of a trainer’s record.
You probably know yourself, or have read elsewhere, that trainers have their own unique strategies for getting winners and it pays to know what these are so that you can build this information into betting systems.
Some trainers are highly-skilled at manipulating a horse’s handicap mark. These trainers are well worth following when it comes to running a horse in a handicap race for the first time. It also pays to know which trainers turn out an above average number of winners when they run a horse after a lengthy layoff, or when they are giving a two-year-old its debut at a certain track.
However, a system that I have found to work particularly well is based on those flat trainers that have an uncanny ability to get a two-year-old to win a race, and are then able to get it to win again on its next outing.
This system has a sound basis because when one considers only those horses that won their last race it is clear that juveniles are more likely to repeat a win than older horses (see Table 8.1).
109
PROFITABLE BETTING SYSTEMS Overall, in my sample of over 60,000 races, two-year-olds that won on their previous outing went on to win again at a rate of over 18 per cent. In comparison winning three-year-olds repeated a win on their next outing on only 17 occasions out of a hundred (16.7 per cent), and horses aged over six years of age only repeated a win at a rate of around 15 per cent.
I think two-year-olds are more likely to repeat a win than older horses for a variety of reasons.
One of the main reasons is that races for juveniles are mostly non-handicaps. A winning two-year-old therefore doesn’t necessarily have to attempt a followup win in a handicap.
Handicap races are bad news for horses trying to repeat a win as they are usually heavily penalised by the handicapper as they have recently won a race. It is often the case that the official handicapper will raise a horse that won its last race by several pounds, even for a narrow victory. This rise in their handicap mark usually means that the horse will either have to be raised in class or will have to concede weight to other runners in a race. In either circumstance it is more difficult for the animal to win again.
Table 8.1: Horses that won their last race and attempted to win again, by age Age
Winners
Runners
% Win
2
1,689
9,246
18.3
3
3,191
19,080
16.7
4
1,641
10,112
16.2
5
1,062
6,675
15.9
6 plus 1,248
8,516
14.7
All
62,145
16.2
10,079
110
TRAINERS TO FOLLOW Juveniles that won their last race would appear to be worth following on their next outing. However, this would not be a profitable strategy on its own. Indeed, in the last five years if you bet on every two-year-old that had won its last race you would have made 9,246 bets and recorded 1,689 winners. This would have resulted in a loss of just over 15 per cent on turnover. This situation though can be radically transformed if one bets on only those trainers that are particularly skilled in getting a juvenile to repeat a win.
Trainers to follow
I have analysed the results of all turf and all-weather flat races run in Great Britain and Ireland over the last few years. The results are presented in Table 8.2.
Table 8.2: Trainers record with juveniles that won their last race, ordered by % win (flat race seasons) Trainer
Winners Runners % Win Profit/loss (£) Profit/loss (%)
PRESCOTT SIR MARK
27
65
41.5
27.48
TREGONING M P
12
33
36.4
19.91
60.33
7
22
31.8
2.20
10.01
CHARLTON R
12
38
31.6
7.34
19.31
NOSEDA J
17
54
31.5
4.40
8.14
JOHNSTON M
43
153
28.1
9.00
5.88
GOSDEN J H M
18
66
27.3
13.23
20.04
7
26
26.9
7.35
28.28
CANDY H
WACHMAN D HASLAM P C
42.28
7
27
25.9
7.32
27.12
OSBORNE J A
14
57
24.6
25.22
44.24
DUNLOP J L
17
70
24.3
0.16
0.23
KIRK S
7
29
24.1
14.63
50.43
ELSWORTH D R C
5
21
23.8
26.25
125.00
22
94
23.4
40.35
42.93
7
32
21.9
5.00
15.63
EASTERBY T D MOORE J S
111
PROFITABLE BETTING SYSTEMS Trainer
Winners Runners % Win Profit/loss (£) Profit/loss (%)
COLE P F I
10
46
21.7
8.73
FAHEY R A
5
26
19.2
2.35
9.04
BURKE K R
8
42
19.0
8.96
21.33
LITTMODEN N P
8
43
18.6
25.68
59.73
RYAN K A
11
64
17.2
2.68
4.18
SMART B
8
48
16.7
2.88
5.99
14
93
15.1
35.68
38.37
BERRY A
9
61
14.8
4.93
8.08
BRITTAIN C E
3
26
11.5
10.50
40.38
331
1363
24.3
350.50
25.72
MEEHAN B J
Total
18.98
As you can see Sir Mark Prescott heads the Table. He is clearly particularly adept at getting a juvenile to repeat a win because, of the 65 horses that he sent out to do so, a total of 27 returned to the winners’ enclosure. This represents a strike rate of nearly 42 per cent!
This remarkable record is not reflected in the horse’s Starting Price because you would have made a profit of £27.48 to a £1 stake, representing an excellent rate of return of over 42 per cent!
However, Sir Mark Prescott is not the only trainer to follow when it comes to getting a horse to repeat a win, and this is good news for turnover.
The trainers Tregoning, Candy, Charlton, and Noseda all recorded a strike rate in excess of 30 per cent. In other words you would have recorded nearly one win for every three bets if you followed this select group.
A little further down the Table you will also note the high number of runners that Mark Johnson turns out to repeat a win. He has a high rate of success with such animals, producing a strike rate of just over 28 per cent. Johnson
112
TRAINERS TO FOLLOW runners though are highly respected by punters these days and followers of his stable would have netted a modest return of just 9 per cent.
John Gosden does almost as well as Mark Johnson with his juveniles but returns a more handsome profit of over 20 per cent. However, David Elsworth takes the prize for the most profitable trainer to follow because, despite holding a licence for as long as I can remember, and saddling high profile horses like Persian Punch and Desert Orchid, his horses are still allowed to go off at decent odds. This allowed his followers to make a profit of 125 per cent over the research period.
Overall if you backed every horse that had won its last race, and was trained by any one of the trainers listed in Table 8.2, you would have made 1,363 bets and this would have had 331 winners. This represents a very respectable strike rate of just over 24 per cent. This keeps losing runs to a reasonably low level, and because the average price of the winners is fairly decent at odds of 8 to 1, the rate of return is nearly 26 per cent.
For those of you that prefer a more selective approach I would advise that you concentrate your wagers on those trainers that record a success rate in excess of 25 per cent.
The nursery system
A few years ago a good friend of mine gave me a very simple betting strategy that has yielded profits years after year. He said to me “Back every horse that Sir Mark Prescott trains that is running in a nursery for the first time”.
113
PROFITABLE BETTING SYSTEMS The rationale behind this system appealed to me because Sir Mark is such a clever student of the Racing Calendar. He carefully studies the conditions of a race and has a great ability for finding the right sort of race for a horse. He also has a reputation for cleverly manipulating a horse’s handicap mark to give it the greatest chance of winning. However, I was initially wary of backing horses in nurseries because these races are traditionally a graveyard for punters, and many serious punters avoid them like the plague.
Nursery races are handicap races for two-year-olds. In order to qualify for a nursery race a two-year-old needs to be allocated an official handicap mark. This figure then determines the amount of weight that a horse will carry. As a rule two-year-olds need to have raced 3 times before they are able to receive a handicap mark. Alternatively they can earn a rating if they have raced just twice provided that they have won at least once.
The official handicapper will attempt to produce a rating that is an accurate reflection of a horse’s ability. This, though, is problematic when rating the runners in a nursery because, unlike handicap races for older horses, twoyear-olds tend not to have raced more than a handful of times, and many will have run only just enough to meet the entry required for a nursery.
Furthermore, juveniles are maturing throughout the racing season and so the quality of their racecourse performances can vary enormously between races. This makes it difficult to judge their true merit. For example, a stunning performance by one horse may simply reflect the fact that the animal had matured more rapidly than his or her rivals, rather than because the horse was naturally more able. In another race, run much later in the racing season, the same horses could meet again and the result could be completely different.
114
TRAINERS TO FOLLOW
The combination of lack of form and the fact that juveniles mature at different rates throughout a racing season makes it very difficult to judge the true ability of two-year-olds. Thus, when rating the runners in a nursery the official handicapper is pretty much in the dark and he or she is very much guessing when allocating the weights.
Sir Mark Prescott is always one step ahead of the handicapper and he has an uncanny ability to be able to accurately judge the true ability of his horses. I need to choose my words carefully here but let us say that an animal in the good care of Sir Mark may not necessarily show its true ability until it has run in a handicap for the first time.
There are good reasons for this. In order to maximise the chances of winning a handicap a trainer needs to get their horse into the race with the lowest possible weight. A horse that won its first three races by an aggregate of thirty lengths would automatically be given top weight if it were then entered in a handicap. In comparison the horse that finished down the field in its previous races wouldn’t attract much interest from the handicapper, or fellow punters for that matter. It would get into a handicap on a low weight, or in others words it would be weighted to win.
My betting records tell me that my friend was absolutely spot-on with his maxim of “Back every horse that Sir Mark Prescott trains that is running in a nursery for the first time”.
In recent years Sir Mark has enjoyed a fantastic record with his runners in nursery handicaps, returning a strike rate of 38 per cent. Moreover if you
115
PROFITABLE BETTING SYSTEMS staked a pound on every one of his runners you would have made a profit of £39.58. This represents a rate of return of nearly 107 per cent!
The problem with my friend’s system was that it was only based on runners from Sir Mark Prescott’s stable. This only provides a handful of selections per season, and while the rate of return is impressive one needs a greater amount of turnover to make the system worthwhile.
I have analysed the results of all nursery races run in Great Britain and Ireland over the last five years or so, and I have identified those trainers that do particularly well with juveniles running in these races for the first time. I have reported the results of this analysis in Table 8.3.
Table 8.3: All nursery races (juvenile handicaps) run in Great Britain and Ireland, by trainer Trainer
Wins Runners % Win Profit/loss (£) Profit/loss (%)
PRESCOTT SIR MARK
14
37
37.8
39.58
GOSDEN J H M
6
19
31.6
2.50
13.16
CUMANI L M
4
14
28.6
7.00
50.00
106.98
COLE P F I
9
33
27.3
33.41
101.24
PRENDERGAST K
7
26
26.9
35.00
134.62
WELD D K
4
18
22.2
9.50
52.78
BELL M L W
10
46
21.7
24.38
52.99
STOUTE SIR MICHAEL 3
15
20.0
4.50
30.00
MORRISON H
2
11
18.2
12.00
109.09
SPEARING J L
2
11
18.2
6.00
54.55
TREGONING M P
2
11
18.2
4.50
40.91
MOORE G L
2
12
16.7
14.50
120.83
DUNLOP J L
9
54
16.7
4.37
8.08
KING A
2
12
16.7
0.50
4.17
BECKETT R M
5
31
16.1
16.00
51.61
EUSTACE J M P
2
13
15.4
11.00
84.62
116
TRAINERS TO FOLLOW Trainer
Wins Runners % Win Profit/loss (£) Profit/loss (%)
OSBORNE J A
6
40
15.0
8.25
20.63
SMART B
6
44
13.6
23.50
53.41
LITTMODEN N P
5
38
13.2
7.00
18.42
BRITTAIN C E
4
32
12.5
27.50
85.94
BEST J R
3
25
12.0
5.00
20.00
RYAN K A
5
44
11.4
1.00
2.27
BUTLER G A
2
18
11.1
27.00
150.00
JARVIS A P
2
19
10.5
5.00
26.32
BALDING A M
2
19
10.5
1.50
7.89
BLANSHARD M
3
29
10.3
22.50
77.59
GLOVER J A
1
12
8.3
9.00
75.00
ELSWORTH D R C
1
12
8.3
1.00
8.33
WHITAKER R M
1
14
7.1
12.00
85.71
DWYER C A
1
15
6.7
11.00
73.33
MARGARSON G G
1
18
5.6
3.00
16.67
Total
138
825
16.7
462.98
56.12
What is particularly striking about the data in Table 8.3 is that some trainers have a very high strike rate for this kind of race. I say that this is striking because the field size in nurseries tends to be large at around 12 runners a race. Therefore is you simply picked a horse at random you would average a win rate of around 8 per cent. However, a number of trainers in Table 8.3 are recording a win rate well in excess of this. For instance, Prescott, Gosden, Cumani, Cole, Weld, Prendergast, Bell, and Stoute record a win rate in excess of 20 per cent or more.
It is always worth following trainers that record a high ratio of winners to runners because this indicates that they are very competent and it also ensures that, by following their runners, losing runs are modest. In comparison I wouldn’t be too confident about following the runners of
117
PROFITABLE BETTING SYSTEMS Margarson or Dwyer because they record such a low strike rate with their runners, even though profits are high.
I would recommend following those trainers that have a strike rate of 20 per cent or more when running a horse in a nursery for the first time. If you stick to this strategy you can expect to record more than one winner in four and earn a rate of return of around 63 per cent on turnover. Jump trainers to follow
It always pays to carefully analyse the record of all jump race trainers because in my opinion there is a huge difference between the most and least competent, with the divide being far greater than that between flat race trainers.
This probably reflects the fact that jumpers are difficult to train because there are so many different types of race, and different types of horse. The trainer of jumpers therefore needs to be able to identify which horse will be suited by fences and hurdles, their distance requirements (which could range between two to four and a half miles) and their preference for different types of track. Jumpers are also particularly susceptible to injury and it takes a particularly skilful trainer to keep a horse sound throughout a season. My research suggests that certain trainers have their own way of working and some specialise in winning certain types of race with certain types of horse. In the analysis that follows I have reproduced the results of my research.
118
TRAINERS TO FOLLOW Non-handicap chases In Table 8.4 I have reproduced my research into the results of all nonhandicap chases and in particular the record of horses that won their previous race and attempted to follow-up their win on their next outing.
There are five trainers that are particularly skilled at getting a horse to repeat a win in a non-handicap chase. Henretta Knight leads the way. She sent out 71 horses with this characteristic during the study period and managed to get 25 of them to return to the winners’ enclosure on their next outing. This represents a strike rate of over 35 per cent.
Punters do not seem to have recognised Knight’s impressive record with chasers that won their last race and you would have made a profit of nearly twenty pounds if you had bet a pound on every one of her runners. Given that she only sent out 71 runners this provides an excellent rate of return of just nearly 28 per cent.
Michael Hourigan does almost equally well with his runners in Ireland. However, what is remarkable about Hourigan's statistics in this area is that overall he has a poor strike rate. This probably explains why punters generally give his runners the swerve, and why the more astute investor, who only follow his chasers when they are attempting to follow up on a win, would have recorded a rate of return of more than 51 per cent! The trainers Henry Daly, Noel Meade and Willie Mullins also do very well with their last time out winners, but rates of return are less impressive than for Knight and Hourigan. Nevertheless if you had followed each of these five trainers over the study period you would have made a profit of £48 to a £1
119
PROFITABLE BETTING SYSTEMS stake and would have recorded nearly one winner for every three selections (31 per cent).
Table 8.4: Winners attempting to follow-up in non-handicap chases Trainer
Winners Runners % Win Profit/loss (£) Profit/loss (%)
KNIGHT MISS H C
25
71
35.2
19.72
27.78
HOURIGAN M
10
31
32.3
15.85
51.13
DALY H D
14
44
31.8
6.80
15.45
MEADE N
13
46
28.3
3.18
6.92
MULLINS W P
10
40
25.0
2.73
6.82
Total
72
232
31.0
48.28
20.81
Non-handicap hurdles
There are a number of trainers that have an impressive record with hurdlers that are attempting to repeat a win on their next outing. Richards, Jefferson, O’Neill and Johnson all record a strike rate of 40 per cent or more with this type of runner.
Nicky Richards is the real money-spinner. He records a strike rate of nearly 50 per cent and a rate of return of over 78 per cent! I can tell you that this is a pattern that he tends to repeat season after season, and his runners deserve particular respect. This is evidenced by the fact that his name appears in a number of the Tables that follow and he is quickly building up a reputation that his late great father Gordon would have been proud of. Unfortunately, I fear that other punters will soon recognise the profitability of Richards’ trained runners and this will diminish the rate of return. I am though comforted by the fact that so far they do not seem to have cottoned on!
120
TRAINERS TO FOLLOW Table 8.5 identifies a couple of other highly profitable trainers to follow. Ian Williams does particularly well with his previous hurdle winners and from 60 runners that won their last race he got 23 to repeat their win on their next outing. He doesn’t appear to be a fashionable trainer to follow, despite his impressive record, and backing his runners would have made a decent profit.
Table 8.5: Winners attempting to follow-up in a non-handicap hurdle Trainer
Winners
RICHARDS N G
29
59
49.2
46.2
78.3
JEFFERSON J M
14
33
42.4
3.9
11.9
O`NEILL JONJO
64
156
41.0
12.2
7.8
JOHNSON HOWARD 30
75
40.0
17.1
22.8
WILLIAMS IAN
23
60
38.3
35.8
59.6
HASLAM P C
13
35
37.1
1.8
5.0
KING A
34
94
36.2
23.1
24.6
MURPHY F
13
38
34.2
1.0
2.5
WEBBER P R
12
37
32.4
0.3
0.9
PHILLIPS R T
11
35
31.4
4.3
12.4
SMITH MRS S J
10
43
23.3
6.4
14.9
161.8
23.1
Total
Runners % Win Profit/loss (£) Profit/loss (%)
701
Novice chases
I know that there are many serious punters that avoid novice chases like the plague and would certainly never bet on a horse that was running in a novice chase for the first time. I do not take this view and while statistically speaking a horse is much more likely to fall in a novice chase, particularly in its first few races over fences, this disguises the fact that some trainers excel in this type of event.
121
PROFITABLE BETTING SYSTEMS Certain trainers have their novice chasers so well schooled that they are at a significant advantage when taking on novices from other stables. It is also worth remembering that these same trainers buy young horses that they have identified as having the scope to win chase races. Therefore these horses may look disappointing when running in National Hunt flat races or over hurdles, but when they debut in a chase they represent an entirely different proposition. Consequently the odds of reward are often generous with this type of runner.
Henry Daly heads Table 8.6. This will be no surprise to those of you familiar with the Daly stable. This stable is really only interested in winning chases and most, if not all, of Daly's horses are purchased with a chasing career in mind. This continues the tradition established by Daly's former boss, the incomparable Captain Tim Forster, who I’m sure must have trained a few hurdlers but I certainly can’t remember any of them winning!
The trainers King, Williams and Smith are also well worth following. Philip Hobbs also has a good record but his runners are well supported in the betting market by fellow punters and the rate of return is a bit too low to be worth following. Mick Easterby though has a good record with novice chasers and backing his runners would have netted you a decent profit in recent years.
122
TRAINERS TO FOLLOW Table 8.6: Trainers record with horses having their first run in a novice chase Trainer DALY H D
Winners Runners % Win Profit/loss (£) Profit/loss (%) 8
33
24.2
41.10
124.55 19.29
SMITH MRS S J
25
104
24.0
20.06
WILLIAMS MISS V
12
50
24.0
22.36
44.72
KING A
14
59
23.7
26.95
45.68
HOBBS P J
38
162
23.5
10.32
6.37
SWAN C F
8
35
22.9
3.75
10.71
RICHARDS N G
10
44
22.7
0.18
0.42
EASTERBY M W
7
31
22.6
37.88
122.18
RYALL B J M
7
33
21.2
22.75
68.94
MULLINS W P
17
85
20.0
10.58
12.45
146
636
23.0
195.93
30.81
Total
Trainers to follow in handicap hurdles and chases
Runners that are making their debut in a handicap are always worthy of respect, particularly when they are in the care of one of the more astute jump race trainers.
In Table 8.7 I have analysed the record of trainers that ran a horse in a handicap hurdle or chase for the first time.
As you can see Nicky Richards heads Table 8.7 in terms of winners to runners. As we have seen earlier in this Chapter, his runners are real moneyspinners, especially when they debut in a handicap.
He recorded 17 winners from 67 runners during the study period. This represents an average of more than one winner for every four runners (25.4 per cent). In terms of profitability a profit of £10.54 would have been earned to
123
PROFITABLE BETTING SYSTEMS a £1 level stake. Had you followed his horses during the last five years you would have made a profit of about 16 per cent on all stakes invested.
Table 8.7: Trainers record with horses running for the first time in a handicap hurdle or chase Trainer
Winners Runners % Win Profit/loss (£) Profit/loss (%)
RICHARDS N G
17
67
25.4
10.54
15.73
EASTERBY T D
7
29
24.1
30.25
104.31
EGERTON C R
7
31
22.6
43.58
140.59
HODGES R J
7
32
21.9
41.00
128.13
NICHOLLS P F
40
184
21.7
2.26
1.23
DALY H D
16
82
19.5
21.75
26.52
Total
94
425
22.1
149.39
35.15
Nicky Richards though is not the most profitable trainer to follow. Charlie Egerton leads the way on that front.
Egerton produced 7 winners from 31 runners and you would have made a level stakes profit of £43.58 backing his runners to level stakes. A profit of £43.58 on a total stake of just £31 gives a rate of return of over 140 per cent.
Ron Hodges is not far behind Egerton in terms of profitability and his runners would have definitely been worth following. The same could also be said of runners saddled by Tim Easterby. You would have made a £30 profit to a £1 stake if you had bet on every horse that he ran in a handicap for the first time, giving a rate of return on stakes invested of over 100 per cent.
Henry Daly is another trainer worth following when running a horse in a handicap for the first time. He did not achieve anywhere near the same level of profitability recorded by the likes of Egerton, Hodges or Easterby but he
124
TRAINERS TO FOLLOW recorded a highly respectable rate of return of over 26 per cent during the study period.
However, I am not confident that the slender profit achieved by runners from the stable of Paul Nicholls would be repeated in the future. His runners nowadays always carry strong public support.
Top weights in National Hunt handicaps
Many punters are put off backing a horse if it is the top weight in a race. You will often hear a comment such as “It’s the best horse in the race, but it has too much weight today”. The idea that a relative difference of a few pounds in weight will prevent a fit horse from winning is a complete fallacy.
The statistics clearly state that the top weight is more likely to win than any other horse in the race. This is a fact that can be exploited by the more astute investor, especially when combined with a knowledge of which trainers have the best record with top weights in handicaps.
In Table 8.8 I have provided details of those trainers that are the most profitable to follow when running a horse off top weight in a handicap hurdle or chase.
A number of the same names discussed in relation to Table 8.7 feature in Table 8.8. Richards, Egerton, Nicholls, Hodges and Easterby all made a profit in recent years with their top weights in handicap hurdles and chases. However, there are a number of other trainers that are also well worth following.
125
PROFITABLE BETTING SYSTEMS Miss Lavelle has an excellent record. She heads Table 8.8 with a strike rate of over 30 per cent and also records the highest rate of return at nearly 108 per cent.
Peter Bowen did nearly as well as Lavelle with his top weighted runners. He turned out 62 top weights and a total of 17 returned to the winners’ enclosure. This represents a strike rate of over 27 per cent and when you consider that field sizes in handicaps tend to be large then this is a very respectable percentage. A level stakes profit of £53.30 is again impressive, translating into a rate of return of almost 86 per cent.
Evan Williams and Howard Johnson are other trainers to keep an eye on during the winter months. Williams had 43 top-weighted runners in handicap hurdles and chases and had 11 winners (26 per cent). These winners resulted in a profit of £16.75, and over 43 bets the rate of return works out at nearly 39 per cent.
Howard Johnson did slightly better than Williams. He had more runners; saddling a total of 68 top weights. These resulted in 17 winners and gave a percentage profit of over 41 per cent.
As mentioned earlier Paul Nicholls is not always a trainer to follow because the odds on his runners tend to be on the short side, but with his top weights he recorded a level stakes profit of £35.61. To achieve this profit you would have had to back 273 of his horses in handicap races and would have recorded 62 winners (22.7 per cent).
126
TRAINERS TO FOLLOW Table 8.8: Record of top weights in handicap hurdles and chases, by trainer Trainer
Winners Runners % Win Profit/loss (£) Profit/loss (%)
LAVELLE MISS E
13
43
30.2
46.42
107.95
BOWEN P
17
62
27.4
53.30
85.97
RICHARDS N G
22
84
26.2
13.90
16.55
WILLIAMS EVAN
11
43
25.6
16.75
38.95
HOWARD JOHNSON
17
68
25.0
28.02
41.20
NICHOLLS P F
62
273
22.7
35.61
13.04
7
31
22.6
4.08
13.15
GUEST R C
28
129
21.7
54.58
42.31
TODHUNTER M
11
51
21.6
11.67
22.88
EGERTON C R
7
34
20.6
6.45
18.97
LUNGO L
32
163
19.6
7.67
4.71
BAILEY K C
13
68
19.1
8.25
12.13
HENDERSON N J
HASLAM P C
20
119
16.8
29.46
24.75
HODGES R J
7
44
15.9
20.92
47.54
EASTERBY T D
7
52
13.5
10.25
19.71
NOLAN P
3
35
8.6
2.00
5.71
I prefer to follow those trainers that have a relatively high strike rate of about 20 per cent plus and that also make a reasonable level of profit on total stakes invested. I would define a reasonable level of profit as being a rate of return in excess of 10 per cent.
There are a number of trainers with a strike rate significantly below 20 at this level. For instance, I wouldn’t bother following top weighted runners trained by Nolan (8.6 per cent) because losing runs would be too long and you are very much dependent on finding a long priced winner to compensate for all the losing bets.
127
Chapter 9: Systems for all-weathers All-weather racing was originally established to keep the racing industry ticking over during any prolonged deep freeze.
I recall that during the middle of the 1980s there was a succession of hard winters that abandoned racing for weeks. A decision was therefore made by the racing authorities to establish racing on an artificial surface, and tracks at Lingfield and Southwell were soon opened. Originally flat and jump races were run on the all-weather but a succession of equine fatalities in jump races resulted in flat-only racing.
All-weather tracks, unlike turf, are very hard wearing. This means that more meetings and more races can take place, and this means more revenue for racecourses and for the racing industry. This has led to the opening of more tracks at Wolverhampton, Kempton, Great Leighs and Dundalk. This trend looks set to continue.
The quality of all-weather racing has improved considerably, especially at Lingfield, but generally speaking the racing at an average meeting is poor, with a day’s programme usually consisting of low-grade handicaps and selling and claiming contests. You do get plenty of exciting finishes but the action can get pretty repetitive, and comparisons with greyhound racing are probably not unfair. However, I have always found plenty of profitable angles on the allweather, and this has kept my interest.
SYSTEMS FOR ALL-WEATHERS One of the most profitable angles that I have found to work on the all-weather, and the one I want to discuss in this Chapter, is based on the study of trainers’ records.
Certainly the only sand that most well known flat trainers are likely to frequent each winter is located on the beaches of Barbados. There are though a band of trainers that have foregone their winter holiday and have specialised in training horses to win on the all-weather tracks. You may not get a golden tan hanging around Southwell and Wolverhampton etc, but there are plenty of decent betting opportunities if you follow the right trainers at these courses.
I have studied the record of trainers that had runners on all-weather tracks in recent years. The results of this research are presented in Table 9.1. The Table is ordered by trainers’ win strike rate to highlight the most consistent trainers.
The first thing that struck me when this data appeared from my computer printer was the length of the list. There are a total of 54 trainers that would have proved to be profitable had you followed them throughout the study period.
Peter Chapple-Hyam had an amazing record on the all-weather. In the sample he had 75 runners and recorded 20 winners. This represents a strike rate of nearly 27 per cent. This is the highest recorded of all trainers that had at least 30 runners on the all-weather during the study period.
The strike rate recorded by Chapple-Hyam is particularly impressive given the fact that the average field size in all-weather races is about 12 to 13 runners. This means that you would expect trainers to average a strike rate of around 8
129
PROFITABLE BETTING SYSTEMS per cent.
This means that Chapple-Hyam is about three times more
successful than the average. It does though mean that the trainers Powell, Siddall, Stimpson, Gallagher, Williams and Wingrove can probably be ignored as trainers –to follow because they average a win rate well below 8 per cent.
Chapple-Hyam also made a very decent profit of almost £25 to a £1 level stake on his 75 runners. This gives a rate of return of 33 per cent. This is much better than Noseda who recorded a very impressive strike rate of over one winner in every four runners, but a rate of return of just under 5 per cent.
Michael Jarvis is one of the top turf flat trainers but he is also adept with his runners on the sand. He recorded a strike rate of over 21 per cent, turning out 215 runners and returning 46 to the winners’ enclosure. The really impressive statistic though is that his rate of return was nearly 42 per cent.
Next in the Table comes Marcus Tregoning. This former assistant to Dick Hern really knows his trade and, like Michael Jarvis, he can produce Classic winners on the turf and plenty of winners on the all-weather also. He had 21 winners from 103 runners on the all-weather in this sample. However, a big name trainer is always going to attract plenty of interest from fellow punters and this probably explains why Tregoning runners only netted a return of just under 11 per cent.
Table 9.1: Profitable trainers –to follow on the all-weather Trainer
Wins Runs % Win Profit/loss (£) Profit/loss (%)
CHAPPLE-HYAM P W
20
75
26.7
24.76
NOSEDA J
49
194
25.3
9.01
4.64
JARVIS M A
46
215
21.4
89.84
41.78
TREGONING M P
21
103
20.4
10.94
10.62
BALDING A
12
62
19.4
19.85
32.02
130
33.01
SYSTEMS FOR ALL-WEATHERS Trainer
Wins Runs % Win Profit/loss (£) Profit/loss (%)
WRAGG G
13
68
19.1
10.49
15.43
TURNELL ANDREW
18
99
18.2
119.25
120.45
BUTLER G A
85
469
18.1
18.37
3.92
MACKIE J
14
78
17.9
53.88
69.07
BARRON T D
123
718
17.1
32.49
4.53
MORRISON H
50
296
16.9
7.83
2.65
5
31
16.1
17.00
54.84
ALLAN R SWINBURN W R
21
133
15.8
16.83
12.66
JARVIS W
32
207
15.5
44.59
21.54
WEBBER P R
12
78
15.4
14.00
17.95
COAKLEY D J
21
142
14.8
37.00
26.06
HARRIS P W
23
158
14.6
27.71
17.54
FANSHAWE J R
22
152
14.5
4.72
3.10
O’NEILL E J
18
126
14.3
48.00
38.10 17.86
GRIFFITHS S P
5
35
14.3
6.25
SEMPLE I
52
373
13.9
35.13
9.42
TOWNSLEY MRS P
10
72
13.9
40.00
55.56
JEFFERSON J M MOORE G L ELSWORTH D R C LEE R
7
53
13.2
34.50
65.09
107
818
13.1
25.41
3.11
29
222
13.1
11.87
5.35
7
54
13.0
21.00
38.89
GOLLINGS S
8
63
12.7
9.50
15.08
HALL MISS S E
5
40
12.5
41.50
103.75
29
237
12.2
12.75
5.38
PEARCE J KING N B
5
43
11.6
20.00
46.51
BARKER D W
18
163
11.0
17.25
10.58
ENRIGHT G P
7
64
10.9
25.25
39.45
LIDDIARD MRS S A
5
46
10.9
13.00
28.26
ALSTON E J
34
321
10.6
0.49
0.15
DUFFIELD MRS A
15
143
10.5
17.13
11.98 36.46
DICKIN R
5
48
10.4
17.50
BALDING J
62
611
10.1
49.63
8.12
JONES D HAYDN
44
434
10.1
151.50
34.91
5
56
8.9
13.00
23.21
HERRIES LADY
131
PROFITABLE BETTING SYSTEMS Trainer
Wins Runs % Win Profit/loss (£) Profit/loss (%)
CHAMINGS P R
19
227
8.4
14.00
BLANSHARD M
27
336
8.0
1.75
6.17 0.52
POWELL T E
4
51
7.8
20.00
39.22
SIDDALL MISS L C
3
46
6.5
48.00
104.35
STIMPSON J T
2
35
5.7
80.00
228.57
GALLAGHER J
8
146
5.5
25.88
17.72
WILLIAMS D L
2
43
4.7
25.00
58.14
WINGROVE K G
1
35
2.9
6.00
17.14
Andrew Balding, son of Ian who trained the great Mill Reef during the 1970s, features prominently in the Table. He had few runners but made a respectable profit for his followers with a rate of return of over 30 per cent.
The trainers I have discussed so far are probably more familiar as turf trainers, but are equally effective at training horses to run on the all-weather. It would though be dangerous to think that a good turf trainer will be just as good with his or her runners on the sand. There are plenty of big name trainers that have a terrible record on the surface. Conversely there are a number of trainers that are big names on the all-weather circuit but who have a less impressive record with their turf runners.
The trainers Gerard Butler, Barron, Morrison and Gary Moore turn out hundreds of horses to run on the all-weather and they have enjoyed remarkable success. You will not get rich betting on their runners but as a group they turn in a profit in most years and their runners deserve plenty of respect.
132
SYSTEMS FOR ALL-WEATHERS All-weather debutantes
Profitable betting systems can also be developed around the record of trainers when they race a horse on the surface for the first time, and I have conducted extensive research on the subject. The results are presented in Table 9.2.
Once again Marcus Tregoning features prominently in the Table and he earned his followers a return of nearly 31 per cent with his all-weather debutantes.
Jeremy Noseda is another big name trainer who is not afraid to run his horses on the all-weather and when he runs a horse on the surface for the first time they deserve plenty of respect. This is clear from his record of 19 winners from 85 runners (22 per cent) and a rate of return of over 16 per cent.
Terry Mills is perhaps less well known than the trainers mentioned so far but he clearly knows the time of day when it comes to getting winners on the allweather and 16 winners from 77 runners is impressive (21 per cent). His rate of return is only 14 per cent but well worth having.
Gerard Butler, Swinburn and Michael Jarvis feature next on the list of trainers to follow, and they all record very decent rates of return on their runners.
Butler introduced 158 horses to the all-weather and produced 31 winners (20 per cent). They were mostly returned at decent prices and the rate of return was over 47 per cent. Walter Swinburn did even better, and followers of his runners would have been very happy with their 75 per cent profit.
133
PROFITABLE BETTING SYSTEMS Table 9.2: Profitable trainers to follow when running a horse on the all-weather for the first time Trainer
Wins Runners % Win Profit/loss (£) Profit/loss (%)
TREGONING M P NOSEDA J
13 19
56 85
23.2 22.4
17.22 14.01
30.75 16.48
MILLS T G
16
77
20.8
10.95
14.23
BUTLER G A
31
158
19.6
74.66
47.26
7
40
17.5
30.00
75.00
JARVIS M A
16
92
17.4
23.20
25.22
BARRON T D
15
88
17.0
17.04
19.37
8
48
16.7
55.75
116.15
FANSHAWE J R
15
96
15.6
5.55
5.78
MORRISON H
14
90
15.6
6.48
7.20
SWINBURN W R
O’NEILL E J
COLLINGRIDGE H J
5
33
15.2
10.18
30.83
JARVIS A P
12
82
14.6
48.25
58.84
D’ARCY P W
8
57
14.0
14.88
26.10
ARBUTHNOT D W P
6
46
13.0
4.24
9.22
ALSTON E J
5
39
12.8
41.67
106.84
DUNLOP E A L
16
130
12.3
7.66
5.89
CHANNON M R
18
152
11.8
4.88
3.21
KELLEWAY MISS GAY
5
44
11.4
2.33
5.30
ELSWORTH D R C
7
68
10.3
35.00
51.47 11.60
TOMPKINS M H
7
72
9.7
8.35
10
114
8.8
1.25
1.10
4
47
8.5
95.00
202.13
LITTMODEN N P
11
142
7.7
22.18
15.62
CHAMINGS P R
3
44
6.8
46.00
104.55
BALDING J
3
45
6.7
23.50
52.22
BOWRING S R
2
31
6.5
12.00
38.71
DOW S
4
66
6.1
18.00
27.27
BERRY JOHN
2
34
5.9
4.50
13.24
JONES D HAYDN
3
54
5.6
43.00
79.63
DUFFIELD MRS A
2
39
5.1
10.00
25.64
BLANSHARD M
3
67
4.5
38.00
56.72
WILLIAMS S C
3
77
3.9
42.50
55.19
HILLS J W CARROLL A W
134
SYSTEMS FOR ALL-WEATHERS Trainer
Wins Runners % Win Profit/loss (£) Profit/loss (%)
FFRENCH DAVIS D J S
1
30
3.3
4.00
13.33
BOYLE J R
1
31
3.2
3.00
9.68
HIATT P W
1
31
3.2
3.00
9.68
MULLINEAUX M
1
38
2.6
13.00
34.21
David Barron features prominently in the table and he has a wonderfully consistent record on the all-weather. His dominance of the sport is perhaps less than it was in the early days but he is still able to exploit his experience to maximum effect when he gets the right horse. A record of 15 winners from 88 runners (17 per cent) and a rate of return of over 19 per cent provides ample evidence that this is a stable to follow.
Eoghan O’Neill is another name to look out for when he gives a horse a run of the all-weather for the first time. He recorded a strike rate of nearly 17 per cent and you would have doubled your money had you bet on all of his runners.
135
Chapter 10: Making a profit from negative factors When it comes to studying the racing form it often pays to eliminate those horses with big negative factors against their name. For instance, a horse wearing blinkers or a tongue strap for the first time can often be quickly eliminated because these factors are very big negatives.
In statistical terms horses wearing a tongue strap or blinkers for the first time are 20 to 1 chances at best. The data is clear that if you backed every horse with any one of these negative characteristics you would pick only one winner in twenty (5 per cent) and would make a staggering loss of well over 50 per cent on turnover.
You may be wondering where all of this is leading (me too!) but in punting, as discussed in Chapter two, lateral thinking often pays. I have therefore been researching whether it is possible to make a profit by backing horses that have very strong negative factors against them.
Horses wearing blinkers for the first time
As we found in Chapter six horses wearing blinkers for the first time are worth opposing, and you can make a decent return by laying such favourites on the exchanges. However, there are exceptions to this rule.
A skilled trainer may know better than others when to apply blinkers to a horse, and which horses are most likely to improve as a result.
MAKING A PROFIT
I have conducted research into the record of trainers when they run horses in blinkers for the first time, and the data confirms that there are indeed some trainers that have a very good record with this type of runner.
In Table 10.1 I have identified those trainers that have the best record with runners wearing blinkers for the first time in National Hunt races.
At the top of the Table you will see Chris ‘Rambo’ Grant. He doesn’t have many runners generally but he knows the time of day when it comes to applying a set of blinkers to a horse. When he uses this tactic his runners should be noted because he has a 27 per cent strike rate with them and recorded a level stakes profit of £13.50. This represents a rate of return of over 120 per cent.
A bigger name trainer appears next in the Table. Alan King is well known as a trainer but perhaps less well known for his record with blinkered runners. His record of 10 winners from 46 runners in the last few years translates into a strike rate of nearly 22 per cent. It is worth putting this into context when you consider that generally horses wearing blinkers for the first time have a strike rate of only 5 per cent. Alan King is therefore hitting a strike rate of four times the average for this type of runner. Followers of his runners would be rewarded with a profit of £21.49 or nearly 47 per cent.
Moving further down the list, Gary Moore runs plenty of his horses in blinkers and he has a very good record with them. They have a strike rate of nearly 17 per cent when wearing the headgear for the first time, and despite being a well-known trainer his runners make a return of nearly 30 per cent.
137
PROFITABLE BETTING SYSTEMS Looking at the other trainers in the list, Eddie O’Grady and Dessie Hughes are others to particularly note.
Table 10.1: The record of National Hunt trainers with runners wearing blinkers for the first time Trainer
Runners Winners Win% Profit/loss(£) Profit/loss(%)
C Grant
11
3
27.3
13.50
122.73
A King
46
10
21.7
21.49
46.72
M J P O'Brien
19
4
21.1
0.73
3.84
E J O'Grady
35
7
20.0
11.58
33.09
B D Leavy
11
2
18.2
14.00
127.27
G L Moore
77
13
16.9
22.75
29.55
Daniel O'Connell
12
2
16.7
7.75
64.58
R Lee
12
2
16.7
5.00
41.67
F Doumen
19
3
15.8
5.00
26.32
Andrew Turnell
13
2
15.4
42.00
323.08
Thomas Foley
13
2
15.4
23.00
176.92
B J Llewellyn
27
4
14.8
21.25
78.70
D T Hughes
67
9
13.4
21.25
31.72
C L Tizzard
24
3
12.5
15.50
64.58
L Wells
18
2
11.1
18.00
100.00
D R Gandolfo
18
2
11.1
13.50
75.00
D G McArdle
10
1
10.0
3.00
30.00
S T Lewis
11
1
9.1
15.00
136.36
H D Daly
22
2
9.1
7.75
35.23
N J Hawke
11
1
9.1
2.00
18.18
Mrs S D Williams
12
1
8.3
9.00
75.00
Michael J McDonagh
13
1
7.7
8.00
61.54
M Sheppard
14
1
7.1
3.00
21.43
Miss A M Newton-Smith 15
1
6.7
11.00
73.33
M E Sowersby
31
2
6.5
75.00
241.94
Evan Williams
21
1
4.8
5.00
23.81
138
MAKING A PROFIT Horses wearing a tongue tie for the first time
It was only a few years ago that tongue-ties had to be officially declared when a trainer entered a horse in a race. Therefore before this time punters were left very much in the dark when it came to knowing about this crucial bit of equipment.
The reason a horse wears a tongue tie is, as the name implies, because a horse needs to have its tongue tied down to save it from swallowing it. A horse that swallows its tongue will go out like a light in a race because it will literally run out of puff. The horse isn’t in any danger of suffocating but it will not be able to breathe optimally and so will not be able to run at anything like racing pace.
A tongue-tie should, if a horse has a tendency for swallowing its tongue, be a positive factor. The fact that it does not is intriguing. In my view tongue-ties (like blinkers) are more a sign that a trainer is getting desperate with a horse rather than it indicating a physical problem. The attitude of some trainers seems to be that “well I’ve tried everything I may as well try a tongue tie”. As a result some horses may not actually be wearing a tongue because they have a breathing problem but because they are incredibly poor quality animals. If this hypothesis holds any weight then some trainers should have a good record with tongue-tied runners because they are applying them for the right reasons, and know when they are needed to improve a horse.
In Table 10.2 I have presented my research on trainers’ records with tonguetied runners.
139
PROFITABLE BETTING SYSTEMS One of the most significant things to note is that there are a number of trainers with a strike rate of over 20 per cent with tongue-tied runners. The trainers Ian Williams, E J O'Grady, G L Moore, and Miss H C Knight all have a strike rate of 20 per cent or more.
There are also a group of trainers that show an amazing rate of return with their tongue-tied runners. The trainers that come out best in the profitability stakes are G L Moore, R H Alner, M Todhunter, P R Webber, Eoin Doyle, Evan Williams, and Andrew Slattery. All of these trainers have a rate of return of at least 100 per cent.
Table 10.2: The record of National Hunt trainers with runners wearing a tongue-tie for the first time Trainer
Runners Winners % Win Profit/loss(£) Profit/loss(%)
Ian Williams
27
7
25.9
11.58
42.89
E J O'Grady
24
6
25.0
8.96
37.33
G L Moore
22
5
22.7
36
163.64
Miss H C Knight
24
5
20.8
20.5
85.42
M J P O'Brien
26
5
19.2
20.1
77.31
R H Alner
11
2
18.2
11
100.00
M Todhunter
12
2
16.7
48
400.00
P R Webber
31
5
16.1
50.5
162.90
Eoin Doyle
19
3
15.8
32
168.42
Evan Williams
32
5
15.6
60.63
189.47
J Neville
13
2
15.4
11
84.62
B J Llewellyn
17
2
11.8
3
17.65
N A Twiston-Davies 60
7
11.7
2
3.33
Andrew Slattery
10
1
10.0
16
160.00
C A Murphy
10
1
10.0
7
70.00
V P Donoghue
10
1
10.0
3
30.00
C L Tizzard
21
2
9.5
3
14.29
J W Mullins
11
1
9.1
6
54.55
140
MAKING A PROFIT Trainer
Runners Winners % Win Profit/loss(£) Profit/loss(%)
Miss E C Lavelle
11
1
9.1
6
54.55
B I Case
11
1
9.1
4
36.36
R T Phillips
35
3
8.6
12
34.29
Mrs D A Hamer
12
1
8.3
9
75.00
B D Leavy
12
1
8.3
3
25.00
R H Buckler
14
1
7.1
1
7.14
Michael Hourigan
15
1
6.7
2
13.33
Ivor Kingston
16
1
6.3
5
31.25
W Storey
17
1
5.9
12
70.59
Miss Jane Thomas
17
1
5.9
9
52.94
S J Mahon
41
1
2.4
10
24.39
Conclusion
It generally does not pay to bet horses that have strong negative factors against them, like the presence of first time blinkers or a tongue-tie. These animals are best laid to lose on the exchanges, especially if they are starting as favourites. However, my research shows that there are exceptions, and a decent profit can be made by backing such horses, provided that they are trained by the right trainer.
141
Chapter 11: Betting systems Over the years I have been asked various questions in relation to developing and implementing betting systems. In this final Chapter I want to deal with some of the most common questions.
Question Is there a system that wins 100 per cent of the time? Answer Sid James discovered the only 100 per cent, fool-proof system that I know of. It involved him reading out the runners and riders until Hattie Jacques’ canary tweeted. The one that got the tweet was the selection. That canary won Sid a fortune. Unfortunately this only happened in the film Carry On At Your Convenience! A system that claimed to win every single time wouldn’t be believable. However, if anyone has a system with a 100 per cent strike rate would they please let me know via the publishers! Dorothy Paget was probably the only person who could have had a 100 per cent successful system. Dorothy was a multimillionaire who was heavily involved in racing in the 1930s. She owned the fives times Cheltenham Gold Cup winner Golden Miller among many others, and was a massive gambler. She also lived a nocturnal existence, rising for breakfast at about 9.00pm. I say that she could have had a 100 per cent successful system because she had an arrangement with her bookmaker that allowed her to place bets on the day’s races after they had been run! Nevertheless, she still managed to lose millions.
BETTING SYSTEMS Question I have purchased a system that has been working OK but I would like to bend its rules in order to generate more selections. Can I do this and still make a consistent profit?
Answer The simple answer is NO. Systems that have been carefully developed have rules that have been set after detailed and thorough analysis. These rules may seem arbitrary but they have no doubt been set to give the highest proportion of winners and the highest level of profitability. However, having said all of this a fundamental part of robust system development is to conduct what statisticians call a sensitivity analysis. This form of analysis tests how sensitive a system is to slight changes in each one of its variables. For instance, it would be a concern if it were found that a system lost its profitability if one of its variables was changed slightly. If this were the case it would be said that this variable was highly sensitive to change and one would question whether the variable could be relied upon to predict future winners. Question I have a betting system that appears to perform better on UK racing than on Irish racing. Should this concern me?
Answer It might be a concern. When I develop systems I often test for a difference between Irish and British races. I’m always reassured if I find no real difference in performance between countries. This is reassuring because it tells me that the system is based on fundamental factors that generally apply across countries. Sometimes if you find a difference you need to look at the data more closely. If a system does record a slightly lower proportion of
143
PROFITABLE BETTING SYSTEMS winners in Ireland than in Britain this may simply reflect the fact that on average field sizes are greater in Irish races, especially for chase races. Question Is there a staking plan that can turn a system that loses at level stakes into a profitable one? Answer Until a few years ago, I hadn’t really bothered with staking plans and preferred to bet only to level stakes.
I hadn’t considered staking plans because I had read so many crazy ones in the past. I recall one strategy that insisted that you double your stakes after each loser. This short cut to the poor house wasn’t one I followed but I know of at least one such unfortunate. His system was a simple one. You backed every horse ridden by Lester Piggott and doubled your stake on his next mount after a loser. He told me that it was a licence to print money because Lester rode so many winners, and never went more than a few rides without a winner. I wasn’t convinced and to be fair he wasn’t doing disastrously, until Lester went about a dozen rides without a winner! He didn’t mention the system to me after that.
I’m now definitely of the view that staking plans can help to maximise profits, but you have to have a genuinely winning system for them to work. I think that the reason that staking plans have such a poor reputation is because punters often try to use them to turn fundamentally unsound systems into profitable ones. There seems to be a rule that the worse the system the more radical the staking plan, but as far as I know, there isn’t a staking plan that can turn a losing system into a winning one. I have never, in all my time researching
144
BETTING SYSTEMS betting systems, found a system that showed a loss at level stakes and made a long-term profit when combined with a staking plan.
However, a system that shows a long-term profit to level stakes can prove to be even more profitable when combined with a sensible staking plan. The problem though is finding a system that shows a long-term profit at level stakes. Question Is there a staking that you would recommend?
Answer When I run a system I prefer to set aside a bank and bet a fixed five per cent of the betting bank on all selections. The five per cent proportional stake is used by many leading professional gamblers and is designed to maximise profits and minimises the risk of losing your betting capital from losing runs. Therefore when I said in Chapter two that you should not change your stake when betting on systems I’m actually talking about a level proportional stake. When betting a fixed proportion of your betting bank, the nominal value of the stake will change from bet to bet, depending on the size of the bank, but the five per cent proportion will not change.
The five per cent figure doesn’t come out of thin air. The statistics tell us that a system with an average strike rate of 50 per cent has about a one per cent probability of hitting a losing run of about seven consecutive losers. This means that there is a one per cent chance of losing about a third of your betting capital (35 per cent) in one losing run and that we would have to hit a sequence of 20 consecutive losers to see most of our betting bank wiped out.
145
PROFITABLE BETTING SYSTEMS The odds of this happening with a system giving 50 per cent winners are very, very low.
We could probably bet a bit more than five per cent of our bank but why be greedy? The five per cent figure is a good balance between risk and reward. You might though want to set a lower proportional stake if your system has a low strike rate. Question I would like to start operating a system with an initial betting bank of £1,000. Is this recommended? Answer I believe strongly that you should always set aside a betting bank for operating systems. My advice is always to only bet with money you can afford to lose and to bet to stakes that you are comfortable with. The actual size of the bank is not important provided that it is sufficient to sustain a reasonable losing run. You also need to be cautious about placing large stakes. It is sometimes difficult to get on if you consistently win, even if you only want to bet at Starting Price. I used to work in the betting industry and I know that bets as small as £100 can attract attention if a ‘face’, that is someone known to be a regular winner, places them. Big wagers that are placed indiscreetly can also reduce the odds on offer and I would always recommend a discrete approach and never be too greedy. The key to betting on systems is to achieve a longterm profit. It is not about hitting one big pay out with one big bet.
146
BETTING SYSTEMS Question I have purchased a system and its previous results look impressive but I would like to paper trade the system before investing real money. Is this sensible or would I miss out?
Answer For those that do not know, paper trading is when one follows a system but does not bet actual money on selections until the system has been proven to make a profit during the period of paper trading.
This approach sounds sensible and I certainly never bet with my own money unless I have strong evidence to believe that the system I am following is valid and reliable. However, the problem with paper trading is that the trading usually only takes place over a short period of time, like one week or one month. A decision is then made about the system that may not reflect its longterm profitability because the sample period that was used may have been unusual.
In order to get an idea about the long-term profitability of a system you may need to trade it for at least one year. It is then very frustrating to add up all the profitable and losing bets at the end of the year to realise that you could have made a packet if you had backed the selections with real money.
If you develop your own systems you can get around this problem by developing a system with what is called a split-half sample. In one random half of the data you develop your system. You then test its performance on the ‘unseen’ data in the other half of the sample. If the system shows a profit over both samples, and provided that the samples are large enough, then you can
147
PROFITABLE BETTING SYSTEMS be fairly confident that you have found a genuinely profitable system. I discuss this in more detail in Chapter two. Question Can you develop a profitable system based on the look of a horse?
Answer In Chapter two I stated that systems need to be based around objective variables that can be measured. I made the point that assessing the physical merits of a horse is highly subjective and that different people will arrive at different judgements. However, there have been attempts to systematically study the appearance of racehorses in order to predict performance.
A fascinating study in the Journal of Applied Animal Behaviour Science in 1997 looked at the pre-race behaviour of horses as a predictor of race finishing order.
The authors considered the look of 867 horses entered in 67 races. They recorded 29 variables for each horse, 19 of which concerned pre-race behaviour and appearance. Fourteen variables were recorded in the parade ring while the rest were scored while the horse was cantering to the starting gate.
The results showed that winners tended to look fitter and were more relaxed and losers tended to be more aroused and required greater control by groom and jockey. Arousal could be detected by the elevation of the head, neck and tail. Thus a horse with a high head and neck carriage on the way to the start might be wasting valuable energy or might have a physical problem.
148
BETTING SYSTEMS Interestingly, sweating didn’t seem very important which runs against the conventional wisdom of many racing pundits that sweating is a sign of anxiety. The researchers found that it wasn’t significant on its own but was moderately significant when combined with other negative factors like signs of arousal in the parade ring or on the way to the start.
The authors get full marks for deciding to take their research out of the ivory tower of the University and trying to apply it to the racetrack. Reading between the lines of the article this sounded like a financial disaster but they did find that while pre race behaviour was useless at picking winners it was useful in identifying losers. However, they couldn’t think of a way of profiting from this information. This though was in the day before betting exchanges when you couldn’t lay horses to lose!
Question What do you think of the betting exchanges? Are they a good thing for punters?
Answer The exchanges have been great for punters and there is plenty of research to suggest that you get longer odds on the exchanges. However, the advent of betting exchanges has had a greater impact than simply providing better value to punters. It has transformed punting and bookmaking. To give an example, I was at the races the other day with a few mates and we were cheering home a 20 to 1 winner. This wasn’t because we had backed it, but because we had layed the short priced favourite to large stakes on the exchanges in the morning. This was possible because the exchanges have allowed punters to become bookmakers, and turned bookmakers into punters. You can change your role from race to race, or play both roles at the same time. There are
149
PROFITABLE BETTING SYSTEMS simply so many more opportunities to make money from betting since the advent of the exchanges. If you have not already got an account get one! Question I have developed a couple of systems that don’t require me to study form. Instead I study fluctuations in the betting market and trade runners that are moving up and down in the market. These are what I call trading systems. Do you use trading systems? Answer In Chapter six I described a number of systems that are based around the betting market. Some of these were designed with the betting exchanges in mind. However, these were not trading systems. Trading systems are very different. I don’t really trade on the exchanges but for those of you that want to get involved I’ll explain in a bit more detail what it is all about.
Trading systems can be used on the exchanges. These aim to guarantee a profit regardless of a race’s outcome. This means that you back and play a horse in a race. Provided that the odds and your stakes are right you should guarantee yourself a profit. These systems though are not fool proof because in-the time it takes you to back and lay a horse its odds could have changed so that the odds are no longer favourable and you can’t arbitrage.
Trading is a good thing for the betting exchange providers, and they effectively encourage trading because it brings more liquidity to the market, as large sums are constantly brought in and out of the market. For instance, BetDaq and Betfair allow computer boffins to produce programs that can access something called their API, which is jargon for basically allowing a computer a direct line into the exchange, making it possible to make scores of
150
BETTING SYSTEMS bets within seconds. It also allows one to play computerised betting strategies. However, you don’t need to be able to program a computer to make use of this innovation.
The open minded policy of the exchanges has resulted in some great software that allows less computer-minded types to trade in a fairly sophisticated way. There are lots of products in this area. I’m not going to recommend any particular product because it depends on what type of strategy you want to employ. However, I think the BetAngel software is worth a mention because this company provide a free product called Bet Angel Basic that enhances the Betfair website interface. The software allows an automatic refresh of odds and features, a ‘green-up’ option that is a useful tool for guaranteeing that you make a profit out of a race, regardless of the outcome, provided that you have a successful trade (i.e. a profit from backing and laying the same horse in a race).
One of the most important tips when trading is knowing when to get out of the market and to take a loss on the chin. The most successful trader will be the one who knows when to get out and quickly recovers their composure to make their next trade. The trader who attempts to chase losses should read Nick Leeson’s Rogue Trader and conclude that this isn’t a good idea!
Question I’ve been making loads of money by playing the in-running markets on the exchanges. I don’t follow any system. I simply back the horse that I think is going to win just before the horse reaches the finish. I often get layed at short odds and I find it is easy to pick up a few quid provided that I can get a big bet into the market before it closes.
151
PROFITABLE BETTING SYSTEMS Answer I know of some big punters that mainly play the in-running markets. They are similar to you in that they like to back big at very short odds on horses that look certain to win. However, you need to remember Devon Loch and Dayjur! They lost despite looking 1000 to 1 on to win just yards before the post. I have played the in-running markets but to be honest I haven’t really made it work for me. Certainly I’ve had success. I once got into the mindset that I had discovered the key to the mint. I simply did what you are doing. I bet big on what looked like near certainties and must have landed scores of successful bets. However, you only need one or two mistakes to undo all your good work if you are backing 1 to 20, 1 to 50 or even 1 to 100 shots! When betting in running I think the most important thing to know is thyself. I know that I’m a greedy fellow and so I can come unstuck when I’m winning. For instance, when I started to make what looked like easy money I couldn’t get enough of it. Thus, it simply wasn’t good enough for me to back winners or to lay near certain losers in-play. I had to back the winners and the placed horses in every race as well. This isn’t a good plan because the level of concentration required simply isn’t sustainable. It was only a matter of time before I made a mistake and after a few reverses I realised that this form of betting wasn’t for me. I prefer to bet on systems. Betting-in-running isn’t really a system and you need to be careful if you are playing to high stakes.
Question I have come across a system that advises that you should not bet selections on a Saturday. Is this a sensible rule? Answer I would have serious reservations about a system that advised that you only bet selections on certain days of the week! A good and reliable system, which
152
BETTING SYSTEMS is based on a set of valid and logical factors, should work irrespective of whether bets are made on a Saturday or any other day of the week. Certainly most systems come up with more selections at weekends than on weekdays because there is simply more meetings at weekends than at other times of the week. I do though know of some academic research that has indicated that you are more likely to be able to strike a value bet on a Saturday and on public holidays. This research shows that in some instances the odds offered by bookmakers are greater than the probability of the bet being successful (in other words the odds represent value). These occasions are more frequent on Saturdays and on bank holidays when there are more ‘mug’ punters playing the game, who stake money on horses that statistically have little chance of winning. This lengthens the odds on the form horses.
The key point I would make is that in order to operate the system successfully you need to back each and every selection, day in day out. This is the only way to avoid the inevitable frustration that follows when you miss a long priced winner. Question I have developed a system based on a horse’s colour. I only bet on greys when they start favourite. Answer Believe it or not a number of academics have researched the performance of racehorses of various colours. In the Journal of Livestock Science earlier this year a group of eggheads explored the hypothesis that grey-coated horses are inferior to non-greys which, if proved true, wouldn’t be good news for your ‘grey-favourite system’.
153
PROFITABLE BETTING SYSTEMS The researchers considered the performances of nearly 2,000 horses over more than 14,000 starts. The results are not actually earth shattering and are more of interest than practical value.
The research found no difference in performance between greys and nongreys but they did find some minor differences in the non-greys, with the darker coloured horses performing slightly better in some respects. I might want to think about this the next time I back a chestnut coloured loser! More seriously, I wouldn’t bother researching systems based on a horse’s coat colour. You will win a prize for originality but probably nothing else! Question I have a couple of very good systems that win consistently, but my wife won’t let me bet. Could I employ an agent? Answer Yes you could employ an agent and I think I have seen these sorts of services advertised on the web. Your predicament though sounds very similar to a friend of mine. He managed to sort it out. But divorce lawyers don’t come cheap!
Question Are there any books on betting that you would recommend? Answer At the end of this book I have added a bibliography of my favourite racing books. Not all of them are about betting, but I have learnt a lot from them.
154
BETTING SYSTEMS Question I’ve had three consecutive losing seasons betting on the horses. I’m desperate. Can you please help? Answer At least you are keeping betting records! I’m always amazed that so many punters don’t bother to keep a record of their betting activities. You need to keep track of your profit and loss and try to identify whether there are any systems that you are playing that aren’t working, or whether you are making any mistakes. For instance, I recall reviewing a fellow punter’s accounts and couldn’t work out why he was losing because he was playing a couple of really good systems that had a very high strike rate. It then dawned on me that the guy wasn’t betting to a consistent stake. His staking was pretty random and he’d been very unlucky in having big bets on losers and smaller stakes on the winners. Once he bet a level stake he was back on track. Question Can a good system become a victim of its own success?
Answer It is certainly true that previously successful systems can suddenly become unprofitable because too many people start to follow them. Commercially available systems in particular do run the risk that they could become victims of their own success. This is less likely to happen if you develop and run your own systems provided that you keep them secret. I’m probably biting my own nose off by writing this book!
155
PROFITABLE BETTING SYSTEMS
Concluding remarks
Developing and betting on systems has been a life long interest for me. I hope that in the proceeding Chapters I have been able to convey some of my enthusiasm, and have at the same time passed on some information that will help you to develop your own systems.
Betting on systems can be enormously satisfying. I still get a buzz when I discover some new fact about horseracing that will give me an edge. I have probably spent more time that is healthy bent over a PC or some dusty back copies of the Racing Post working out new systems. But I don’t regret a second, and I will keep on developing and betting on systems until I join the great racecourse in the sky. I simply can’t afford to stop.
156
Selected Bibliography Ashforth, D (1996). Hitting the turf: a punting life. Headline
Beyer, A (1994). Picking Winners: a horseplayers guide. Houghton Mifflin
Beyer, A (1978). My $50,000 year at the racetrack . Harcourt Brace Jovanovich
Beyer, A (1993). Beyer on Speed. Houghton Mifflin
Beyer, A (1994). The winning horseplayer. Houghton Mifflin
Braddock, P (1983). Braddock’s complete guide to horserace selection and betting. Longman
Grove, W M, & Meehl, P E (1996). Comparative efficiency of informal (subjective, impressionistic) and formal (mechanical, algorithmic) prediction procedures: The clinical statistical controversy. Psychology, Public Policy, and Law, 2, 293–323
Hutson G D and Haskell M (1997). Pre-race behaviour of horses as a predictor of race finishing order. Journal of Applied Animal Behaviour Science. Volume 53, Number 4, July, pp. 231-248(18)
Helm, M (1994). Exploring pedigree: handicapping’s newest frontier. City Miner Books
Holt, C (1994). Profitable winners always back value winners. Fineform Racing Publications
May, P (1998). Forecasting methods for horse racing. Raceform
May, P (2004). Horseracing: a guide to profitable betting. Raceform
Mordin, N (1992). Betting for a Living. Aesculus Press
Mordin, N (1996). Mordin on time. Aesculus Press
Mordin, N (2002). Winning without thinking. Aesculus Press
Potts, A (1995). Against the Crowd: the methods of a modern backer. Aesculus Press
Quirin, W (1979). Winning at the races: computer discoveries in thoroughbred handicapping. William Morrow
Skiena, S (2001). Calculated bets: computers, gambling and mathematical modelling to win. Cambridge University Press
""#* Performance in racehorses of various colours. Livestock Science. Volume 106, Issues 2-3, February 2007, Pages 282-286
Stich, L (2004). Pedigree handicapping. Daily Racing Form Press
Taylor, R (2006). Pace wins the race. Sportsworld Publishing
HIGH STAKES STOCK TITLES 9781843440253 9781901982077 9781843440260 9781843440192 9781843440000 9781843440079 9781843440185 9781843440109 9781843440383 9781843440321 9781843440413 9781843440147 9781843440239 9781843440024 9781843440536 9781843440437 9781843440376 9781843440529 9781843440062 9781843440178 9781843440543 9781843440017 9781843440208 9781843440154
Chance Amir D.Aczel The Money Spinners Jacques Black The Art of Bookmaking Malcolm Boyle Fixed Odds Sports Betting Joe Buchdahl Win At Greyhound Racing H. Edwards Clarke The Science of Winning Burton P. Fabricand Game, Set & Matched Iain Fletcher Football Fortunes Bill Hunter The Smart Money Michael Konik Thirteen Against the Bank Norman Leigh Automatic Exchange Betting Colin McGee The Science of Poker Dr M. Mahmood The Making of a Poker Player Matt Matros Forecasting Methods Horseracing Peter May Cheltenham Profiles 2009 DW Redpath 1001 Great Gambling Tips Graham Sharpe Gent’s Guide Calculating Winning Bets G Sharpe Improve Your Betting SmarterSig Total Poker David Spanier Profitable Football Betting Paul N Steele Profitable Betting Strategies R Taylor The Education of a Poker Player H O Yardley Roulette: Playing to Win Brett Morton Betting to Win Prof. Leighton Vaughan Williams
7.99 7.99 9.99 12.99 9.99 14.99 9.99 9.99 12.99 7.99 40.00 12.99 9.99 14.99 9.99 9.99 7.99 12.99 9.99 9.99 14.99 9.99 12.99 12.99
Please send orders to: High Stakes Bookshop (PBS), 21 Great Ormond St, London WC1N 3JB, Add 15% P&P. T: 0207 4301021 F: 0207 4300021 Cheques payable to High Stakes or pay by credit card (Visa, MC, Maestro) quoting card number, expiry date, 3 digit security code, and Valid from date and Issue number where appropriate.
Order online at www.highstakespublishing.co.uk or www.highstakes.co.uk